Merge branch 'dev' into feature/cog-2734-cognee-feedbacks-interactions-poc-to-prod

.env.template

@@ -6,7 +6,7 @@
 #       Default graph database    : Kuzu
 #
 #       These default databases are all file-based, so no extra setup is needed
-#       for local use.
+#       for local use. The data by default will be stored in your .venv
 ###############################################################################
 
 ################################################################################
@@ -16,7 +16,7 @@
 STRUCTURED_OUTPUT_FRAMEWORK="instructor"
 
 LLM_API_KEY="your_api_key"
-LLM_MODEL="openai/gpt-4o-mini"
+LLM_MODEL="openai/gpt-5-mini"
 LLM_PROVIDER="openai"
 LLM_ENDPOINT=""
 LLM_API_VERSION=""
@@ -33,11 +33,20 @@ EMBEDDING_MAX_TOKENS=8191
 
 # If using BAML structured output these env variables will be used
 BAML_LLM_PROVIDER=openai
-BAML_LLM_MODEL="gpt-4o-mini"
+BAML_LLM_MODEL="gpt-5-mini"
 BAML_LLM_ENDPOINT=""
 BAML_LLM_API_KEY="your_api_key"
 BAML_LLM_API_VERSION=""
 
+
+################################################################################
+#  📂 ROOT DIRECTORY FOR DATABASES
+################################################################################
+# Set up the Cognee system directory. Cognee will store system files and databases here.
+# Useful for setting root directory inside docker and also to avoid storing the databases in .venv
+# DATA_ROOT_DIRECTORY='/Users/<user>/Desktop/cognee/.cognee_data/'
+# SYSTEM_ROOT_DIRECTORY='/Users/<user>/Desktop/cognee/.cognee_system/'
+
 ################################################################################
 # 🗄️ Relational database settings
 ################################################################################
@@ -85,12 +94,7 @@ VECTOR_DB_PROVIDER="lancedb"
 VECTOR_DB_URL=
 VECTOR_DB_KEY=
 
-################################################################################
+
-#  📂 ROOT DIRECTORY IF USING COGNEE LIB INSIDE A DOCKER
-################################################################################
-# Set up the Cognee system directory. Cognee will store system files and databases here.
-# DATA_ROOT_DIRECTORY='/Users/<user>/Desktop/cognee/.cognee_data/'
-# SYSTEM_ROOT_DIRECTORY='/Users/<user>/Desktop/cognee/.cognee_system/'
 
 ################################################################################
 #  🔄  MIGRATION (RELATIONAL → GRAPH) SETTINGS
@@ -157,8 +161,8 @@ LITELLM_LOG="ERROR"
 # Uncomment + fill values to switch.
 
 ########## Azure OpenAI #######################################################
-#LLM_MODEL="azure/gpt-4o-mini"
+#LLM_MODEL="azure/gpt-5-mini"
-#LLM_ENDPOINT="https://DNS.azure.com/openai/deployments/gpt-4o-mini"
+#LLM_ENDPOINT="https://DNS.azure.com/openai/deployments/gpt-5-mini"
 #LLM_API_KEY="<<TALK TO YOUR AZURE GUY"
 #LLM_API_VERSION="2024-12-01-preview"
 

.github/actions/cognee_setup/action.yml

@@ -1,11 +1,15 @@
 name: cognee-setup
-description: "Sets up Python, installs Poetry, loads venv from cache, and installs dependencies for Cognee."
+description: "Sets up Python, installs uv, and installs dependencies for Cognee."
 
 inputs:
   python-version:
     description: "Which Python version to use"
     required: false
     default: "3.11.x"
+  extra-dependencies:
+    description: "Additional extra dependencies to install (space-separated)"
+    required: false
+    default: ""
 
 runs:
   using: "composite"
@@ -16,18 +20,25 @@ runs:
       with:
         python-version: ${{ inputs.python-version }}
 
-    - name: Install Poetry
+    - name: Install uv
-      shell: bash
+      uses: astral-sh/setup-uv@v4
-      run: |
+      with:
-        python -m pip install --upgrade pip
+        enable-cache: true
-        pip install poetry
     
-    - name: Rebuild Poetry lock file
+    - name: Rebuild uv lockfile
       shell: bash
       run: |
-        rm poetry.lock
+        rm uv.lock
-        poetry lock
+        uv lock
 
     - name: Install dependencies
       shell: bash
-      run: poetry install --no-interaction -E api -E docs -E evals -E gemini -E codegraph -E ollama -E dev -E neo4j
+      run: |
+        EXTRA_ARGS=""
+        if [ -n "${{ inputs.extra-dependencies }}" ]; then
+          IFS=' ' read -r -a deps <<< "${{ inputs.extra-dependencies }}"
+          for extra in "${deps[@]}"; do
+            EXTRA_ARGS="$EXTRA_ARGS --extra $extra"
+          done
+        fi
+        uv sync --extra api --extra docs --extra evals --extra gemini --extra codegraph --extra ollama --extra dev --extra neo4j $EXTRA_ARGS

.github/actions/setup_neo4j/action.yml

@@ -0,0 +1,67 @@
+name: 'Setup Neo4j with Graph Data Science'
+description: 'Sets up a Neo4j instance with APOC and Graph Data Science plugins for testing'
+inputs:
+  neo4j-version:
+    description: 'Neo4j version to use'
+    required: false
+    default: '5.21'
+  neo4j-password:
+    description: 'Password for Neo4j'
+    required: false
+    default: 'cognee_test_password'
+outputs:
+  neo4j-url:
+    description: 'Neo4j connection URL'
+    value: 'bolt://localhost:7687'
+  neo4j-username:
+    description: 'Neo4j username'
+    value: 'neo4j'
+  neo4j-password:
+    description: 'Neo4j password'
+    value: ${{ inputs.neo4j-password }}
+runs:
+  using: 'composite'
+  steps:
+    - name: Start Neo4j with GDS
+      shell: bash
+      run: |
+        docker run -d \
+          --name neo4j-test \
+          -p 7474:7474 -p 7687:7687 \
+          -e NEO4J_AUTH="neo4j/${{ inputs.neo4j-password }}" \
+          -e NEO4J_PLUGINS='["apoc", "graph-data-science"]' \
+          -e NEO4J_dbms_security_procedures_unrestricted="apoc.*,gds.*" \
+          -e NEO4J_apoc_export_file_enabled=true \
+          -e NEO4J_apoc_import_file_enabled=true \
+          neo4j:${{ inputs.neo4j-version }}
+    
+    - name: Wait for Neo4j to be ready
+      shell: bash
+      run: |
+        echo "Waiting for Neo4j to start..."
+        timeout=60
+        counter=0
+        
+        while [ $counter -lt $timeout ]; do
+          if docker exec neo4j-test cypher-shell -u neo4j -p "${{ inputs.neo4j-password }}" "RETURN 1" > /dev/null 2>&1; then
+            echo "Neo4j is ready!"
+            break
+          fi
+          echo "Waiting... ($counter/$timeout)"
+          sleep 2
+          counter=$((counter + 2))
+        done
+        
+        if [ $counter -ge $timeout ]; then
+          echo "Neo4j failed to start within $timeout seconds"
+          docker logs neo4j-test
+          exit 1
+        fi
+    
+    - name: Verify GDS is available
+      shell: bash
+      run: |
+        echo "Verifying Graph Data Science library is available..."
+        docker exec neo4j-test cypher-shell -u neo4j -p "${{ inputs.neo4j-password }}" \
+          "CALL gds.version() YIELD gdsVersion RETURN gdsVersion"
+        echo "GDS verification complete!"

.github/workflows/basic_tests.yml

@@ -98,7 +98,7 @@ jobs:
           python-version: ${{ inputs.python-version }}
 
       - name: Run Unit Tests
-        run: poetry run pytest cognee/tests/unit/
+        run: uv run pytest cognee/tests/unit/
 
   integration-tests:
     name: Run Integration Tests
@@ -115,7 +115,7 @@ jobs:
           python-version: ${{ inputs.python-version }}
 
       - name: Run Integration Tests
-        run: poetry run pytest cognee/tests/integration/
+        run: uv run pytest cognee/tests/integration/
 
   simple-examples:
     name: Run Simple Examples
@@ -144,7 +144,7 @@ jobs:
           python-version: ${{ inputs.python-version }}
 
       - name: Run Simple Examples
-        run: poetry run python ./examples/python/simple_example.py
+        run: uv run python ./examples/python/simple_example.py
 
   simple-examples-baml:
     name: Run Simple Examples BAML
@@ -180,7 +180,7 @@ jobs:
           python-version: ${{ inputs.python-version }}
 
       - name: Run Simple Examples
-        run: poetry run python ./examples/python/simple_example.py
+        run: uv run python ./examples/python/simple_example.py
 
   graph-tests:
     name: Run Basic Graph Tests
@@ -209,4 +209,4 @@ jobs:
           python-version: ${{ inputs.python-version }}
 
       - name: Run Graph Tests
-        run: poetry run python ./examples/python/code_graph_example.py --repo_path ./cognee/tasks/graph
+        run: uv run python ./examples/python/code_graph_example.py --repo_path ./cognee/tasks/graph

.github/workflows/db_examples_tests.yml

@@ -54,10 +54,6 @@ jobs:
         with:
           python-version: ${{ inputs.python-version }}
 
-      - name: Install Neo4j extra
-        run: |
-          poetry install -E neo4j
-
       - name: Run Neo4j Example
         env:
           ENV: dev
@@ -74,7 +70,7 @@ jobs:
           GRAPH_DATABASE_USERNAME: "neo4j"
           GRAPH_DATABASE_PASSWORD: ${{ secrets.NEO4J_API_KEY }}
         run: |
-          poetry run python examples/database_examples/neo4j_example.py
+          uv run python examples/database_examples/neo4j_example.py
 
   run-db-example-kuzu:
     name: "Kuzu DB Example Test"
@@ -90,9 +86,8 @@ jobs:
         with:
           python-version: ${{ inputs.python-version }}
 
-      - name: Install Kuzu extra
+      - name: Dependencies already installed
-        run: |
+        run: echo "Dependencies already installed in setup"
-          poetry install
 
       - name: Run Kuzu Example
         env:
@@ -107,7 +102,7 @@ jobs:
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
           GRAPH_DATABASE_PROVIDER: "kuzu"
         run: |
-          poetry run python examples/database_examples/kuzu_example.py
+          uv run python examples/database_examples/kuzu_example.py
 
   run-db-example-pgvector:
     name: "PostgreSQL PGVector DB Example Test"
@@ -138,10 +133,7 @@ jobs:
         uses: ./.github/actions/cognee_setup
         with:
           python-version: ${{ inputs.python-version }}
-
+          extra-dependencies: "postgres"
-      - name: Install PGVector extra
-        run: |
-          poetry install -E postgres
 
       - name: Run PGVector Example
         env:
@@ -155,4 +147,4 @@ jobs:
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
         run: |
-          poetry run python examples/database_examples/pgvector_example.py
+          uv run python examples/database_examples/pgvector_example.py

.github/workflows/e2e_tests.yml

@@ -57,7 +57,7 @@ jobs:
           EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
-        run: poetry run python ./cognee/tests/test_cognee_server_start.py
+        run: uv run python ./cognee/tests/test_cognee_server_start.py
 
   run-telemetry-test:
     name: Run Telemetry Test
@@ -81,7 +81,7 @@ jobs:
           EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
-        run: poetry run python ./cognee/tests/test_telemetry.py
+        run: uv run python ./cognee/tests/test_telemetry.py
 
   run-telemetry-pipeline-test:
     name: Run Telemetry Pipeline Test
@@ -110,7 +110,7 @@ jobs:
           EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
-        run: poetry run python ./cognee/tests/test_library.py
+        run: uv run python ./cognee/tests/test_library.py
 
   run-deduplication-test:
     name: Deduplication Test
@@ -140,10 +140,7 @@ jobs:
         uses: ./.github/actions/cognee_setup
         with:
           python-version: '3.11.x'
-
+          extra-dependencies: "postgres"
-      - name: Install specific db dependency
-        run: |
-          poetry install -E postgres
 
       - name: Run Deduplication Example
         env:
@@ -154,7 +151,7 @@ jobs:
           EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
-        run: poetry run python ./cognee/tests/test_deduplication.py
+        run: uv run python ./cognee/tests/test_deduplication.py
 
   run-deletion-test:
     name: Deletion Test
@@ -179,7 +176,7 @@ jobs:
           EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
-        run: poetry run python ./cognee/tests/test_deletion.py
+        run: uv run python ./cognee/tests/test_deletion.py
 
   run-s3-bucket-test:
       name: S3 Bucket Test
@@ -192,10 +189,10 @@ jobs:
           uses: ./.github/actions/cognee_setup
           with:
             python-version: '3.11.x'
+            extra-dependencies: "aws"
 
-        - name: Install specific S3 dependency
+        - name: Dependencies already installed
-          run: |
+          run: echo "Dependencies already installed in setup"
-            poetry install -E aws
 
         - name: Run S3 Bucket Test
           env:
@@ -210,7 +207,7 @@ jobs:
             EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
             AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_ID }}
             AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
-          run: poetry run python ./cognee/tests/test_s3.py
+          run: uv run python ./cognee/tests/test_s3.py
 
   test-parallel-databases:
     name: Test using different async databases in parallel in Cognee
@@ -224,9 +221,8 @@ jobs:
         with:
           python-version: '3.11.x'
 
-      - name: Install specific graph db dependency
+      - name: Dependencies already installed
-        run: |
+        run: echo "Dependencies already installed in setup"
-          poetry install
 
       - name: Run parallel databases test
         env:
@@ -239,7 +235,7 @@ jobs:
           EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
-        run: poetry run python ./cognee/tests/test_parallel_databases.py
+        run: uv run python ./cognee/tests/test_parallel_databases.py
 
   test-permissions:
     name: Test permissions with different situations in Cognee
@@ -253,9 +249,8 @@ jobs:
         with:
           python-version: '3.11.x'
 
-      - name: Install specific graph db dependency
+      - name: Dependencies already installed
-        run: |
+        run: echo "Dependencies already installed in setup"
-          poetry install
 
       - name: Run parallel databases test
         env:
@@ -268,7 +263,7 @@ jobs:
           EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
-        run: poetry run python ./cognee/tests/test_permissions.py
+        run: uv run python ./cognee/tests/test_permissions.py
 
   test-graph-edges:
     name: Test graph edge ingestion
@@ -282,9 +277,8 @@ jobs:
         with:
           python-version: '3.11.x'
 
-      - name: Install specific graph db dependency
+      - name: Dependencies already installed
-        run: |
+        run: echo "Dependencies already installed in setup"
-          poetry install
 
       - name: Run graph edges test
         env:
@@ -297,4 +291,4 @@ jobs:
           EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
-        run: poetry run python ./cognee/tests/test_edge_ingestion.py
+        run: uv run python ./cognee/tests/test_edge_ingestion.py

.github/workflows/examples_tests.yml

@@ -20,7 +20,7 @@ jobs:
         env:
           LLM_API_KEY: ${{ secrets.OPENAI_API_KEY }}
           OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
-        run: poetry run python ./examples/python/multimedia_example.py
+        run: uv run python ./examples/python/multimedia_example.py
 
   test-eval-example:
     name: Run Eval Example
@@ -33,10 +33,7 @@ jobs:
         uses: ./.github/actions/cognee_setup
         with:
           python-version: '3.11.x'
-
+          extra-dependencies: "deepeval"
-      - name: Install specific eval dependency
-        run: |
-          poetry install -E deepeval
 
       - name: Run Evaluation Framework Example
         env:
@@ -49,7 +46,7 @@ jobs:
           EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
-        run: poetry run python ./cognee/eval_framework/run_eval.py
+        run: uv run python ./cognee/eval_framework/run_eval.py
 
   test-descriptive-metrics:
     name: Run Descriptive Metrics Example
@@ -63,6 +60,10 @@ jobs:
         with:
           python-version: '3.11.x'
 
+      - name: Setup Neo4j with GDS
+        uses: ./.github/actions/setup_neo4j
+        id: neo4j
+
       - name: Run Descriptive Graph Metrics Example
         env:
           LLM_MODEL: ${{ secrets.LLM_MODEL }}
@@ -74,7 +75,11 @@ jobs:
           EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
-        run: poetry run python ./cognee/tests/tasks/descriptive_metrics/networkx_metrics_test.py
+          GRAPH_DATABASE_PROVIDER: "neo4j"
+          GRAPH_DATABASE_URL: ${{ steps.neo4j.outputs.neo4j-url }}
+          GRAPH_DATABASE_USERNAME: ${{ steps.neo4j.outputs.neo4j-username }}
+          GRAPH_DATABASE_PASSWORD: ${{ steps.neo4j.outputs.neo4j-password }}
+        run: uv run python ./cognee/tests/tasks/descriptive_metrics/neo4j_metrics_test.py
 
 
   test-dynamic-steps-metrics:
@@ -100,4 +105,4 @@ jobs:
           EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
-        run: poetry run python ./examples/python/dynamic_steps_example.py
+        run: uv run python ./examples/python/dynamic_steps_example.py

.github/workflows/graph_db_tests.yml

@@ -28,9 +28,8 @@ jobs:
         with:
           python-version: ${{ inputs.python-version }}
 
-      - name: Install specific db dependency
+      - name: Dependencies already installed
-        run: |
+        run: echo "Dependencies already installed in setup"
-          poetry install
 
       - name: Run Kuzu Tests
         env:
@@ -43,7 +42,7 @@ jobs:
           EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
-        run: poetry run python ./cognee/tests/test_kuzu.py
+        run: uv run python ./cognee/tests/test_kuzu.py
 
       - name: Run Weighted Edges Tests with Kuzu
         env:
@@ -57,7 +56,7 @@ jobs:
           EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
-        run: poetry run pytest cognee/tests/unit/interfaces/graph/test_weighted_edges.py -v
+        run: uv run pytest cognee/tests/unit/interfaces/graph/test_weighted_edges.py -v
 
   run-neo4j-tests:
     name: Neo4j Tests
@@ -72,10 +71,6 @@ jobs:
         with:
           python-version: ${{ inputs.python-version }}
 
-      - name: Install specific db dependency
-        run: |
-          poetry install -E neo4j
-
       - name: Run default Neo4j
         env:
           ENV: 'dev'
@@ -91,7 +86,7 @@ jobs:
           GRAPH_DATABASE_URL: ${{ secrets.NEO4J_API_URL }}
           GRAPH_DATABASE_PASSWORD: ${{ secrets.NEO4J_API_KEY }}
           GRAPH_DATABASE_USERNAME: "neo4j"
-        run: poetry run python ./cognee/tests/test_neo4j.py
+        run: uv run python ./cognee/tests/test_neo4j.py
 
       - name: Run Weighted Edges Tests with Neo4j
         env:
@@ -108,4 +103,4 @@ jobs:
           EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
-        run: poetry run pytest cognee/tests/unit/interfaces/graph/test_weighted_edges.py -v
+        run: uv run pytest cognee/tests/unit/interfaces/graph/test_weighted_edges.py -v

.github/workflows/python_version_tests.yml

@@ -55,7 +55,7 @@ jobs:
 
       - name: Run unit tests
         shell: bash
-        run: poetry run pytest cognee/tests/unit/
+        run: uv run pytest cognee/tests/unit/
         env:
           PYTHONUTF8: 1
           LLM_PROVIDER: openai
@@ -73,7 +73,7 @@ jobs:
       - name: Run integration tests
         if: ${{ !contains(matrix.os, 'windows') }}
         shell: bash
-        run: poetry run pytest cognee/tests/integration/
+        run: uv run pytest cognee/tests/integration/
         env:
           PYTHONUTF8: 1
           LLM_PROVIDER: openai
@@ -103,11 +103,11 @@ jobs:
           EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
-        run: poetry run python ./cognee/tests/test_library.py
+        run: uv run python ./cognee/tests/test_library.py
 
-      - name: Build with Poetry
+      - name: Build with uv
         shell: bash
-        run: poetry build
+        run: uv build
 
       - name: Install Package
         if: ${{ !contains(matrix.os, 'windows') }}

.github/workflows/relational_db_migration_tests.yml

@@ -58,10 +58,10 @@ jobs:
         uses: ./.github/actions/cognee_setup
         with:
           python-version: '3.11.x'
+          extra-dependencies: "postgres"
 
       - name: Install specific db dependency
-        run: |
+        run: echo "Dependencies already installed in setup"
-          poetry install -E postgres
 
       - name: Run PostgreSQL Script to create test data (Chinook_PostgreSql.sql)
         env:
@@ -90,7 +90,7 @@ jobs:
           EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
-        run: poetry run python ./cognee/tests/test_relational_db_migration.py
+        run: uv run python ./cognee/tests/test_relational_db_migration.py
 
   run-relational-db-migration-test-kuzu:
     name: Kuzu Relational DB Migration Test
@@ -120,10 +120,10 @@ jobs:
         uses: ./.github/actions/cognee_setup
         with:
           python-version: '3.11.x'
+          extra-dependencies: "postgres"
 
       - name: Install specific db dependency
-        run: |
+        run: echo "Dependencies already installed in setup"
-          poetry install -E postgres
 
       - name: Run PostgreSQL Script to create test data (Chinook_PostgreSql.sql)
         env:
@@ -154,7 +154,7 @@ jobs:
           EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
-        run: poetry run python ./cognee/tests/test_relational_db_migration.py
+        run: uv run python ./cognee/tests/test_relational_db_migration.py
 
   run-relational-db-migration-test-neo4j:
     name: Neo4j Relational DB Migration Test
@@ -184,10 +184,10 @@ jobs:
         uses: ./.github/actions/cognee_setup
         with:
           python-version: '3.11.x'
+          extra-dependencies: "postgres"
 
       - name: Install specific db dependency
-        run: |
+        run: echo "Dependencies already installed in setup"
-          poetry install -E postgres -E neo4j
 
       - name: Run PostgreSQL Script to create test data (Chinook_PostgreSql.sql)
         env:
@@ -221,4 +221,4 @@ jobs:
           EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
-        run: poetry run python ./cognee/tests/test_relational_db_migration.py
+        run: uv run python ./cognee/tests/test_relational_db_migration.py

.github/workflows/reusable_notebook.yml

@@ -46,10 +46,7 @@ jobs:
         uses: ./.github/actions/cognee_setup
         with:
           python-version: ${{ inputs.python-version }}
-
+          extra-dependencies: "notebook"
-      - name: Install specific db dependency
-        run: |
-          poetry install -E notebook
 
       - name: Execute Jupyter Notebook
         env:
@@ -64,7 +61,7 @@ jobs:
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
         run: |
-          poetry run jupyter nbconvert \
+          uv run jupyter nbconvert \
           --to notebook \
           --execute ${{ inputs.notebook-location }} \
           --output executed_notebook.ipynb \

.github/workflows/search_db_tests.yml

@@ -28,9 +28,8 @@ jobs:
         with:
           python-version: ${{ inputs.python-version }}
 
-      - name: Install specific db dependency
+      - name: Dependencies already installed
-        run: |
+        run: echo "Dependencies already installed in setup"
-          poetry install
 
       - name: Run Kuzu search Tests
         env:
@@ -46,7 +45,7 @@ jobs:
           GRAPH_DATABASE_PROVIDER: 'kuzu'
           VECTOR_DB_PROVIDER: 'lancedb'
           DB_PROVIDER: 'sqlite'
-        run: poetry run python ./cognee/tests/test_search_db.py
+        run: uv run python ./cognee/tests/test_search_db.py
 
   run-neo4j-lance-sqlite-search-tests:
     name: Search test for Neo4j/LanceDB/Sqlite
@@ -78,9 +77,8 @@ jobs:
         with:
           python-version: ${{ inputs.python-version }}
 
-      - name: Install specific db dependency
+      - name: Dependencies already installed
-        run: |
+        run: echo "Dependencies already installed in setup"
-          poetry install -E neo4j
 
       - name: Run Neo4j search Tests
         env:
@@ -99,7 +97,7 @@ jobs:
           GRAPH_DATABASE_URL: bolt://localhost:7687
           GRAPH_DATABASE_USERNAME: neo4j
           GRAPH_DATABASE_PASSWORD: pleaseletmein
-        run: poetry run python ./cognee/tests/test_search_db.py
+        run: uv run python ./cognee/tests/test_search_db.py
 
   run-kuzu-pgvector-postgres-search-tests:
       name: Search test for Kuzu/PGVector/Postgres
@@ -129,9 +127,10 @@ jobs:
           uses: ./.github/actions/cognee_setup
           with:
             python-version: ${{ inputs.python-version }}
+            extra-dependencies: "postgres"
 
-        - name: Install dependencies
+        - name: Dependencies already installed
-          run: poetry install -E postgres
+          run: echo "Dependencies already installed in setup"
 
         - name: Run Kuzu/PGVector/Postgres Tests
           env:
@@ -152,7 +151,7 @@ jobs:
             DB_PORT: 5432
             DB_USERNAME: cognee
             DB_PASSWORD: cognee
-          run: poetry run python ./cognee/tests/test_search_db.py
+          run: uv run python ./cognee/tests/test_search_db.py
 
   run-neo4j-pgvector-postgres-search-tests:
     name: Search test for Neo4j/PGVector/Postgres
@@ -195,10 +194,10 @@ jobs:
         uses: ./.github/actions/cognee_setup
         with:
           python-version: ${{ inputs.python-version }}
+          extra-dependencies: "postgres"
 
-      - name: Install dependencies
+      - name: Dependencies already installed
-        run: |
+        run: echo "Dependencies already installed in setup"
-          poetry install -E neo4j -E postgres
 
       - name: Run Neo4j + PGVector + Postgres search Tests
         env:
@@ -222,4 +221,4 @@ jobs:
           DB_PORT: 5432
           DB_USERNAME: cognee
           DB_PASSWORD: cognee
-        run: poetry run python ./cognee/tests/test_search_db.py
+        run: uv run python ./cognee/tests/test_search_db.py

.github/workflows/test_gemini.yml

@@ -26,4 +26,4 @@ jobs:
           EMBEDDING_MODEL: "gemini/text-embedding-004"
           EMBEDDING_DIMENSIONS: "768"
           EMBEDDING_MAX_TOKENS: "8076"
-        run: poetry run python ./examples/python/simple_example.py
+        run: uv run python ./examples/python/simple_example.py

.github/workflows/test_ollama.yml

@@ -26,7 +26,7 @@ jobs:
 
       - name: Install torch dependency
         run: |
-          poetry add torch
+          uv add torch
 
 #      - name: Install ollama
 #        run: curl -fsSL https://ollama.com/install.sh | sh
@@ -101,4 +101,4 @@ jobs:
           EMBEDDING_ENDPOINT: "http://localhost:11434/api/embeddings"
           EMBEDDING_DIMENSIONS: "4096"
           HUGGINGFACE_TOKENIZER: "Salesforce/SFR-Embedding-Mistral"
-        run: poetry run python ./examples/python/simple_example.py
+        run: uv run python ./examples/python/simple_example.py

.github/workflows/test_s3_file_storage.yml

@@ -36,4 +36,4 @@ jobs:
           EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
-        run: poetry run python ./cognee/tests/test_s3_file_storage.py
+        run: uv run python ./cognee/tests/test_s3_file_storage.py

.github/workflows/vector_db_tests.yml

@@ -56,7 +56,7 @@ jobs:
   #         EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
   #         EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
   #         EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
-  #       run: poetry run python ./cognee/tests/test_chromadb.py
+  #       run: uv run python ./cognee/tests/test_chromadb.py
 
 
   run-postgres-tests:
@@ -87,10 +87,7 @@ jobs:
         uses: ./.github/actions/cognee_setup
         with:
           python-version: ${{ inputs.python-version }}
-
+          extra-dependencies: "postgres"
-      - name: Install specific db dependency
-        run: |
-          poetry install -E postgres
 
       - name: Run PGVector Tests
         env:
@@ -103,4 +100,4 @@ jobs:
           EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
-        run: poetry run python ./cognee/tests/test_pgvector.py
+        run: uv run python ./cognee/tests/test_pgvector.py

.github/workflows/weighted_edges_tests.yml

@@ -31,7 +31,7 @@ jobs:
         python-version: ['3.11', '3.12']
     env:
       LLM_PROVIDER: openai
-      LLM_MODEL: gpt-4o-mini
+      LLM_MODEL: gpt-5-mini
       LLM_API_KEY: ${{ secrets.LLM_API_KEY }}
 
     steps:
@@ -47,11 +47,11 @@ jobs:
 
       - name: Run Weighted Edges Unit Tests
         run: |
-          poetry run pytest cognee/tests/unit/interfaces/graph/test_weighted_edges.py -v --tb=short
+          uv run pytest cognee/tests/unit/interfaces/graph/test_weighted_edges.py -v --tb=short
 
       - name: Run Standard Graph Tests (Regression)
         run: |
-          poetry run pytest cognee/tests/unit/interfaces/graph/get_graph_from_model_unit_test.py -v --tb=short
+          uv run pytest cognee/tests/unit/interfaces/graph/get_graph_from_model_unit_test.py -v --tb=short
 
   test-with-different-databases:
     name: Test Weighted Edges with Different Graph Databases
@@ -61,14 +61,12 @@ jobs:
         database: ['kuzu', 'neo4j']
         include:
           - database: kuzu
-            install_extra: ""
             graph_db_provider: "kuzu"
           - database: neo4j
-            install_extra: "-E neo4j"
             graph_db_provider: "neo4j"
     env:
       LLM_PROVIDER: openai
-      LLM_MODEL: gpt-4o-mini
+      LLM_MODEL: gpt-5-mini
       LLM_ENDPOINT: https://api.openai.com/v1/
       LLM_API_KEY: ${{ secrets.LLM_API_KEY }}
       LLM_API_VERSION: "2024-02-01"
@@ -88,22 +86,21 @@ jobs:
         with:
           python-version: '3.11'
 
-      - name: Install Database Dependencies
+      - name: Dependencies already installed
-        run: |
+        run: echo "Dependencies already installed in setup"
-          poetry install ${{ matrix.install_extra }}
 
       - name: Run Weighted Edges Tests
         env:
           GRAPH_DATABASE_PROVIDER: ${{ matrix.graph_db_provider }}
         run: |
-          poetry run pytest cognee/tests/unit/interfaces/graph/test_weighted_edges.py -v --tb=short
+          uv run pytest cognee/tests/unit/interfaces/graph/test_weighted_edges.py -v --tb=short
 
   test-examples:
     name: Test Weighted Edges Examples
     runs-on: ubuntu-22.04
     env:
       LLM_PROVIDER: openai
-      LLM_MODEL: gpt-4o-mini
+      LLM_MODEL: gpt-5-mini
       LLM_ENDPOINT: https://api.openai.com/v1/
       LLM_API_KEY: ${{ secrets.LLM_API_KEY }}
       LLM_API_VERSION: "2024-02-01"
@@ -125,7 +122,7 @@ jobs:
 
       - name: Test Weighted Edges Example
         run: |
-          poetry run python examples/python/weighted_edges_example.py
+          uv run python examples/python/weighted_edges_example.py
 
       - name: Verify Visualization File Created
         run: |

README.md

@@ -92,7 +92,8 @@ Your contributions are at the core of making this a true open source project. An
 ## 📦 Installation
 
 You can install Cognee using either **pip**, **poetry**, **uv** or any other python package manager.
-Cognee supports Python 3.8 to 3.12
+
+Cognee supports Python 3.10 to 3.13
 
 ### With pip
 
@@ -102,7 +103,7 @@ pip install cognee
 
 ## Local Cognee installation
 
-You can install the local Cognee repo using **pip**, **poetry** and **uv**.
+You can install the local Cognee repo using **uv**, **pip** and **poetry**.
 For local pip installation please make sure your pip version is above version 21.3.
 
 ### with UV with all optional dependencies

cognee-frontend/src/ui/Partials/SettingsModal/Settings.tsx

@@ -34,8 +34,8 @@
 // };
 
 // const defaultModel = {
-//   label: 'gpt-4o-mini',
+//   label: 'gpt-5-mini',
-//   value: 'gpt-4o-mini',
+//   value: 'gpt-5-mini',
 // };
 
 // export default function Settings({ onDone = () => {}, submitButtonText = 'Save' }) {

cognee-mcp/pyproject.toml

@@ -8,7 +8,7 @@ requires-python = ">=3.10"
 dependencies = [
     # For local cognee repo usage remove comment bellow and add absolute path to cognee. Then run `uv sync --reinstall` in the mcp folder on local cognee changes.
 #    "cognee[postgres,codegraph,gemini,huggingface,docs,neo4j] @ file:/Users/vasilije/Projects/tiktok/cognee",
-    "cognee[postgres,codegraph,gemini,huggingface,docs,neo4j]==0.2.1",
+    "cognee[postgres,codegraph,gemini,huggingface,docs,neo4j]==0.2.2",
     "fastmcp>=2.10.0,<3.0.0",
     "mcp>=1.12.0,<2.0.0",
     "uv>=0.6.3,<1.0.0",

cognee-mcp/src/server.py

@@ -221,14 +221,6 @@ async def cognify(data: str, graph_model_file: str = None, graph_model_name: str
     - The actual cognify process may take significant time depending on text length
     - Use the cognify_status tool to check the progress of the operation
 
-    Raises
-    ------
-    InvalidValueError
-        If LLM_API_KEY is not set
-    ValueError
-        If chunks exceed max token limits (reduce chunk_size)
-    DatabaseNotCreatedError
-        If databases are not properly initialized
     """
 
     async def cognify_task(
@@ -512,14 +504,6 @@ async def search(search_query: str, search_type: str) -> list:
     - Different search types produce different output formats
     - The function handles the conversion between Cognee's internal result format and MCP's output format
 
-    Raises
-    ------
-    InvalidValueError
-        If LLM_API_KEY is not set (for LLM-based search types)
-    ValueError
-        If query_text is empty or search parameters are invalid
-    NoDataError
-        If no relevant data found for the search query
     """
 
     async def search_task(search_query: str, search_type: str) -> str:

cognee-starter-kit/.env.template

@@ -1,6 +1,6 @@
 # In case you choose to use OpenAI provider, just adjust the model and api_key.
 LLM_API_KEY=""
-LLM_MODEL="openai/gpt-4o-mini"
+LLM_MODEL="openai/gpt-5-mini"
 LLM_PROVIDER="openai"
 # Not needed if you use OpenAI
 LLM_ENDPOINT=""

cognee/api/v1/add/add.py

@@ -128,17 +128,12 @@ async def add(
 
         Optional:
         - LLM_PROVIDER: "openai" (default), "anthropic", "gemini", "ollama"
-        - LLM_MODEL: Model name (default: "gpt-4o-mini")
+        - LLM_MODEL: Model name (default: "gpt-5-mini")
         - DEFAULT_USER_EMAIL: Custom default user email
         - DEFAULT_USER_PASSWORD: Custom default user password
         - VECTOR_DB_PROVIDER: "lancedb" (default), "chromadb", "pgvector"
-        - GRAPH_DATABASE_PROVIDER: "kuzu" (default), "neo4j", "networkx"
+        - GRAPH_DATABASE_PROVIDER: "kuzu" (default), "neo4j"
 
-    Raises:
-        FileNotFoundError: If specified file paths don't exist
-        PermissionError: If user lacks access to files or dataset
-        UnsupportedFileTypeError: If file format cannot be processed
-        InvalidValueError: If LLM_API_KEY is not set or invalid
     """
     tasks = [
         Task(resolve_data_directories, include_subdirectories=True),

cognee/api/v1/cognify/cognify.py

@@ -91,7 +91,7 @@ async def cognify(
                 - LangchainChunker: Recursive character splitting with overlap
                 Determines how documents are segmented for processing.
         chunk_size: Maximum tokens per chunk. Auto-calculated based on LLM if None.
-                   Formula: min(embedding_max_tokens, llm_max_tokens // 2)
+                   Formula: min(embedding_max_completion_tokens, llm_max_completion_tokens // 2)
                    Default limits: ~512-8192 tokens depending on models.
                    Smaller chunks = more granular but potentially fragmented knowledge.
         ontology_file_path: Path to RDF/OWL ontology file for domain-specific entity types.
@@ -177,14 +177,6 @@ async def cognify(
         - LLM_PROVIDER, LLM_MODEL, VECTOR_DB_PROVIDER, GRAPH_DATABASE_PROVIDER
         - LLM_RATE_LIMIT_ENABLED: Enable rate limiting (default: False)
         - LLM_RATE_LIMIT_REQUESTS: Max requests per interval (default: 60)
-
-    Raises:
-        DatasetNotFoundError: If specified datasets don't exist
-        PermissionError: If user lacks processing rights
-        InvalidValueError: If LLM_API_KEY is not set
-        OntologyParsingError: If ontology file is malformed
-        ValueError: If chunks exceed max token limits (reduce chunk_size)
-        DatabaseNotCreatedError: If databases are not properly initialized
     """
     tasks = await get_default_tasks(user, graph_model, chunker, chunk_size, ontology_file_path)
 

cognee/api/v1/config/config.py

@@ -2,7 +2,6 @@
 
 import os
 from cognee.base_config import get_base_config
-from cognee.exceptions import InvalidValueError, InvalidAttributeError
 from cognee.modules.cognify.config import get_cognify_config
 from cognee.infrastructure.data.chunking.config import get_chunk_config
 from cognee.infrastructure.databases.vector import get_vectordb_config
@@ -11,6 +10,7 @@ from cognee.infrastructure.llm.config import (
     get_llm_config,
 )
 from cognee.infrastructure.databases.relational import get_relational_config, get_migration_config
+from cognee.api.v1.exceptions.exceptions import InvalidConfigAttributeError
 
 
 class config:
@@ -92,9 +92,7 @@ class config:
             if hasattr(llm_config, key):
                 object.__setattr__(llm_config, key, value)
             else:
-                raise InvalidAttributeError(
+                raise InvalidConfigAttributeError(attribute=key)
-                    message=f"'{key}' is not a valid attribute of the config."
-                )
 
     @staticmethod
     def set_chunk_strategy(chunk_strategy: object):
@@ -131,9 +129,7 @@ class config:
             if hasattr(relational_db_config, key):
                 object.__setattr__(relational_db_config, key, value)
             else:
-                raise InvalidAttributeError(
+                raise InvalidConfigAttributeError(attribute=key)
-                    message=f"'{key}' is not a valid attribute of the config."
-                )
 
     @staticmethod
     def set_migration_db_config(config_dict: dict):
@@ -145,9 +141,7 @@ class config:
             if hasattr(migration_db_config, key):
                 object.__setattr__(migration_db_config, key, value)
             else:
-                raise InvalidAttributeError(
+                raise InvalidConfigAttributeError(attribute=key)
-                    message=f"'{key}' is not a valid attribute of the config."
-                )
 
     @staticmethod
     def set_graph_db_config(config_dict: dict) -> None:
@@ -171,9 +165,7 @@ class config:
             if hasattr(vector_db_config, key):
                 object.__setattr__(vector_db_config, key, value)
             else:
-                raise InvalidAttributeError(
+                InvalidConfigAttributeError(attribute=key)
-                    message=f"'{key}' is not a valid attribute of the config."
-                )
 
     @staticmethod
     def set_vector_db_key(db_key: str):

cognee/api/v1/datasets/routers/get_datasets_router.py

@@ -13,7 +13,7 @@ from cognee.infrastructure.databases.relational import get_relational_engine
 from cognee.modules.data.methods import get_authorized_existing_datasets
 from cognee.modules.data.methods import create_dataset, get_datasets_by_name
 from cognee.shared.logging_utils import get_logger
-from cognee.api.v1.delete.exceptions import DataNotFoundError, DatasetNotFoundError
+from cognee.api.v1.exceptions import DataNotFoundError, DatasetNotFoundError
 from cognee.modules.users.models import User
 from cognee.modules.users.methods import get_authenticated_user
 from cognee.modules.users.permissions.methods import (

cognee/api/v1/delete/delete.py

@@ -16,7 +16,7 @@ from cognee.modules.users.methods import get_default_user
 from cognee.modules.data.methods import get_authorized_existing_datasets
 from cognee.context_global_variables import set_database_global_context_variables
 
-from cognee.api.v1.delete.exceptions import (
+from cognee.api.v1.exceptions import (
     DocumentNotFoundError,
     DatasetNotFoundError,
     DocumentSubgraphNotFoundError,

cognee/api/v1/exceptions/__init__.py

@@ -0,0 +1,13 @@
+"""
+Custom exceptions for the Cognee API.
+
+This module defines a set of exceptions for handling various data errors
+"""
+
+from .exceptions import (
+    InvalidConfigAttributeError,
+    DocumentNotFoundError,
+    DatasetNotFoundError,
+    DataNotFoundError,
+    DocumentSubgraphNotFoundError,
+)

cognee/api/v1/exceptions/exceptions.py

@@ -1,10 +1,19 @@
-from cognee.exceptions import CogneeApiError
+from cognee.exceptions import CogneeConfigurationError, CogneeValidationError
 from fastapi import status
 
 
-class DocumentNotFoundError(CogneeApiError):
+class InvalidConfigAttributeError(CogneeConfigurationError):
-    """Raised when a document cannot be found in the database."""
+    def __init__(
+        self,
+        attribute: str,
+        name: str = "InvalidConfigAttributeError",
+        status_code: int = status.HTTP_400_BAD_REQUEST,
+    ):
+        message = f"'{attribute}' is not a valid attribute of the configuration."
+        super().__init__(message, name, status_code)
 
+
+class DocumentNotFoundError(CogneeValidationError):
     def __init__(
         self,
         message: str = "Document not found in database.",
@@ -14,9 +23,7 @@ class DocumentNotFoundError(CogneeApiError):
         super().__init__(message, name, status_code)
 
 
-class DatasetNotFoundError(CogneeApiError):
+class DatasetNotFoundError(CogneeValidationError):
-    """Raised when a dataset cannot be found."""
-
     def __init__(
         self,
         message: str = "Dataset not found.",
@@ -26,9 +33,7 @@ class DatasetNotFoundError(CogneeApiError):
         super().__init__(message, name, status_code)
 
 
-class DataNotFoundError(CogneeApiError):
+class DataNotFoundError(CogneeValidationError):
-    """Raised when a dataset cannot be found."""
-
     def __init__(
         self,
         message: str = "Data not found.",
@@ -38,9 +43,7 @@ class DataNotFoundError(CogneeApiError):
         super().__init__(message, name, status_code)
 
 
-class DocumentSubgraphNotFoundError(CogneeApiError):
+class DocumentSubgraphNotFoundError(CogneeValidationError):
-    """Raised when a document's subgraph cannot be found in the graph database."""
-
     def __init__(
         self,
         message: str = "Document subgraph not found in graph database.",

cognee/api/v1/responses/models.py

@@ -70,7 +70,7 @@ class ResponseRequest(InDTO):
     tool_choice: Optional[Union[str, Dict[str, Any]]] = "auto"
     user: Optional[str] = None
     temperature: Optional[float] = 1.0
-    max_tokens: Optional[int] = None
+    max_completion_tokens: Optional[int] = None
 
 
 class ToolCallOutput(BaseModel):

cognee/api/v1/search/search.py

@@ -161,13 +161,6 @@ async def search(
         - VECTOR_DB_PROVIDER: Must match what was used during cognify
         - GRAPH_DATABASE_PROVIDER: Must match what was used during cognify
 
-    Raises:
-        DatasetNotFoundError: If specified datasets don't exist or aren't accessible
-        PermissionDeniedError: If user lacks read access to requested datasets
-        NoDataError: If no relevant data found for the search query
-        InvalidValueError: If LLM_API_KEY is not set (for LLM-based search types)
-        ValueError: If query_text is empty or search parameters are invalid
-        CollectionNotFoundError: If vector collection not found (data not processed)
     """
     # We use lists from now on for datasets
     if isinstance(datasets, UUID) or isinstance(datasets, str):

cognee/eval_framework/eval_config.py

@@ -25,7 +25,7 @@ class EvalConfig(BaseSettings):
         "EM",
         "f1",
     ]  # Use only 'correctness' for DirectLLM
-    deepeval_model: str = "gpt-4o-mini"
+    deepeval_model: str = "gpt-5-mini"
 
     # Metrics params
     calculate_metrics: bool = True

cognee/exceptions/__init__.py

@@ -2,13 +2,13 @@
 Custom exceptions for the Cognee API.
 
 This module defines a set of exceptions for handling various application errors,
-such as service failures, resource conflicts, and invalid operations.
+such as System, Validation, Configuration or TransientErrors
 """
 
 from .exceptions import (
     CogneeApiError,
-    ServiceError,
+    CogneeSystemError,
-    InvalidValueError,
+    CogneeValidationError,
-    InvalidAttributeError,
+    CogneeConfigurationError,
-    CriticalError,
+    CogneeTransientError,
 )

cognee/exceptions/exceptions.py

@@ -35,37 +35,57 @@ class CogneeApiError(Exception):
         return f"{self.name}: {self.message} (Status code: {self.status_code})"
 
 
-class ServiceError(CogneeApiError):
+class CogneeSystemError(CogneeApiError):
-    """Failures in external services or APIs, like a database or a third-party service"""
+    """System error"""
 
     def __init__(
         self,
-        message: str = "Service is unavailable.",
+        message: str = "A system error occurred.",
-        name: str = "ServiceError",
+        name: str = "CogneeSystemError",
-        status_code=status.HTTP_422_UNPROCESSABLE_ENTITY,
+        status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
+        log=True,
+        log_level="ERROR",
     ):
-        super().__init__(message, name, status_code)
+        super().__init__(message, name, status_code, log, log_level)
 
 
-class InvalidValueError(CogneeApiError):
+class CogneeValidationError(CogneeApiError):
+    """Validation error"""
+
     def __init__(
         self,
-        message: str = "Invalid Value.",
+        message: str = "A validation error occurred.",
-        name: str = "InvalidValueError",
+        name: str = "CogneeValidationError",
         status_code=status.HTTP_422_UNPROCESSABLE_ENTITY,
+        log=True,
+        log_level="ERROR",
     ):
-        super().__init__(message, name, status_code)
+        super().__init__(message, name, status_code, log, log_level)
 
 
-class InvalidAttributeError(CogneeApiError):
+class CogneeConfigurationError(CogneeApiError):
+    """SystemConfigError"""
+
     def __init__(
         self,
-        message: str = "Invalid attribute.",
+        message: str = "A system configuration error occurred.",
-        name: str = "InvalidAttributeError",
+        name: str = "CogneeConfigurationError",
-        status_code=status.HTTP_422_UNPROCESSABLE_ENTITY,
+        status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
+        log=True,
+        log_level="ERROR",
     ):
-        super().__init__(message, name, status_code)
+        super().__init__(message, name, status_code, log, log_level)
 
 
-class CriticalError(CogneeApiError):
+class CogneeTransientError(CogneeApiError):
-    pass
+    """TransientError"""
+
+    def __init__(
+        self,
+        message: str = "A transient error occurred.",
+        name: str = "CogneeTransientError",
+        status_code=status.HTTP_503_SERVICE_UNAVAILABLE,
+        log=True,
+        log_level="ERROR",
+    ):
+        super().__init__(message, name, status_code, log, log_level)

cognee/infrastructure/data/exceptions/__init__.py

@@ -0,0 +1,7 @@
+"""
+Custom exceptions for the Cognee API.
+
+This module defines a set of exceptions for handling various data errors
+"""
+
+from .exceptions import KeywordExtractionError

cognee/infrastructure/data/exceptions/exceptions.py

@@ -0,0 +1,22 @@
+from cognee.exceptions import (
+    CogneeValidationError,
+)
+from fastapi import status
+
+
+class KeywordExtractionError(CogneeValidationError):
+    """
+    Raised when a provided value is syntactically valid but semantically unacceptable
+    for the given operation.
+
+    Example:
+        - Passing an empty string to a keyword extraction function.
+    """
+
+    def __init__(
+        self,
+        message: str = "Extract_keywords cannot extract keywords from empty text.",
+        name: str = "KeywordExtractionError",
+        status_code: int = status.HTTP_400_BAD_REQUEST,
+    ):
+        super().__init__(message, name, status_code)

cognee/infrastructure/data/utils/extract_keywords.py

@@ -1,6 +1,6 @@
 from sklearn.feature_extraction.text import TfidfVectorizer
 
-from cognee.exceptions import InvalidValueError
+from cognee.infrastructure.data.exceptions.exceptions import KeywordExtractionError
 from cognee.shared.utils import extract_pos_tags
 
 
@@ -8,7 +8,7 @@ def extract_keywords(text: str) -> list[str]:
     """
     Extract keywords from the provided text string.
 
-    This function raises an InvalidValueError if the input text is empty. It processes the
+    This function raises an KeyWordExtractionError if the input text is empty. It processes the
     text to extract parts of speech, focusing on nouns, and uses TF-IDF to identify the most
     relevant keywords based on their frequency. The function returns a list of up to 15
     keywords, each having more than 3 characters.
@@ -25,7 +25,7 @@ def extract_keywords(text: str) -> list[str]:
           with more than 3 characters.
     """
     if len(text) == 0:
-        raise InvalidValueError(message="extract_keywords cannot extract keywords from empty text.")
+        raise KeywordExtractionError()
 
     tags = extract_pos_tags(text)
     nouns = [word for (word, tag) in tags if tag == "NN"]

cognee/infrastructure/databases/exceptions/EmbeddingException.py

@@ -1,20 +0,0 @@
-from cognee.exceptions import CogneeApiError
-from fastapi import status
-
-
-class EmbeddingException(CogneeApiError):
-    """
-    Custom exception for handling embedding-related errors.
-
-    This exception class is designed to indicate issues specifically related to embeddings
-    within the application. It extends the base exception class CogneeApiError and allows
-    for customization of the error message, name, and status code.
-    """
-
-    def __init__(
-        self,
-        message: str = "Embedding Exception.",
-        name: str = "EmbeddingException",
-        status_code=status.HTTP_422_UNPROCESSABLE_ENTITY,
-    ):
-        super().__init__(message, name, status_code)

cognee/infrastructure/databases/exceptions/__init__.py

@@ -8,4 +8,7 @@ from .exceptions import (
     EntityNotFoundError,
     EntityAlreadyExistsError,
     DatabaseNotCreatedError,
+    EmbeddingException,
+    MissingQueryParameterError,
+    MutuallyExclusiveQueryParametersError,
 )

cognee/infrastructure/databases/exceptions/exceptions.py

@@ -1,13 +1,13 @@
 from fastapi import status
-from cognee.exceptions import CogneeApiError, CriticalError
+from cognee.exceptions import CogneeSystemError, CogneeValidationError, CogneeConfigurationError
 
 
-class DatabaseNotCreatedError(CriticalError):
+class DatabaseNotCreatedError(CogneeSystemError):
     """
     Represents an error indicating that the database has not been created. This error should
     be raised when an attempt is made to access the database before it has been initialized.
 
-    Inherits from CriticalError. Overrides the constructor to include a default message and
+    Inherits from CogneeSystemError. Overrides the constructor to include a default message and
     status code.
     """
 
@@ -20,10 +20,10 @@ class DatabaseNotCreatedError(CriticalError):
         super().__init__(message, name, status_code)
 
 
-class EntityNotFoundError(CogneeApiError):
+class EntityNotFoundError(CogneeValidationError):
     """
     Represents an error when a requested entity is not found in the database. This class
-    inherits from CogneeApiError.
+    inherits from CogneeValidationError.
 
     Public methods:
 
@@ -49,11 +49,11 @@ class EntityNotFoundError(CogneeApiError):
         # super().__init__(message, name, status_code) :TODO: This is not an error anymore with the dynamic exception handling therefore we shouldn't log error
 
 
-class EntityAlreadyExistsError(CogneeApiError):
+class EntityAlreadyExistsError(CogneeValidationError):
     """
     Represents an error when an entity creation is attempted but the entity already exists.
 
-    This class is derived from CogneeApiError and is used to signal a conflict in operations
+    This class is derived from CogneeValidationError and is used to signal a conflict in operations
     involving resource creation.
     """
 
@@ -66,11 +66,11 @@ class EntityAlreadyExistsError(CogneeApiError):
         super().__init__(message, name, status_code)
 
 
-class NodesetFilterNotSupportedError(CogneeApiError):
+class NodesetFilterNotSupportedError(CogneeConfigurationError):
     """
     Raise an exception when a nodeset filter is not supported by the current database.
 
-    This exception inherits from `CogneeApiError` and is designed to provide information
+    This exception inherits from `CogneeConfigurationError` and is designed to provide information
     about the specific issue of unsupported nodeset filters in the context of graph
     databases.
     """
@@ -84,3 +84,51 @@ class NodesetFilterNotSupportedError(CogneeApiError):
         self.message = message
         self.name = name
         self.status_code = status_code
+
+
+class EmbeddingException(CogneeConfigurationError):
+    """
+    Custom exception for handling embedding-related errors.
+
+    This exception class is designed to indicate issues specifically related to embeddings
+    within the application. It extends the base exception class CogneeConfigurationError allows
+    for customization of the error message, name, and status code.
+    """
+
+    def __init__(
+        self,
+        message: str = "Embedding Exception.",
+        name: str = "EmbeddingException",
+        status_code=status.HTTP_422_UNPROCESSABLE_ENTITY,
+    ):
+        super().__init__(message, name, status_code)
+
+
+class MissingQueryParameterError(CogneeValidationError):
+    """
+    Raised when neither 'query_text' nor 'query_vector' is provided,
+    and at least one is required to perform the operation.
+    """
+
+    def __init__(
+        self,
+        name: str = "MissingQueryParameterError",
+        status_code: int = status.HTTP_400_BAD_REQUEST,
+    ):
+        message = "One of query_text or query_vector must be provided!"
+        super().__init__(message, name, status_code)
+
+
+class MutuallyExclusiveQueryParametersError(CogneeValidationError):
+    """
+    Raised when both 'text' and 'embedding' are provided to the search function,
+    but only one type of input is allowed at a time.
+    """
+
+    def __init__(
+        self,
+        name: str = "MutuallyExclusiveQueryParametersError",
+        status_code: int = status.HTTP_400_BAD_REQUEST,
+    ):
+        message = "The search function accepts either text or embedding as input, but not both."
+        super().__init__(message, name, status_code)

cognee/infrastructure/databases/graph/get_graph_engine.py

@@ -21,10 +21,6 @@ async def get_graph_engine() -> GraphDBInterface:
     if hasattr(graph_client, "initialize"):
         await graph_client.initialize()
 
-    # Handle loading of graph for NetworkX
-    if config["graph_database_provider"].lower() == "networkx" and graph_client.graph is None:
-        await graph_client.load_graph_from_file()
-
     return graph_client
 
 
@@ -181,8 +177,7 @@ def create_graph_engine(
             graph_id=graph_identifier,
         )
 
-    from .networkx.adapter import NetworkXAdapter
+    raise EnvironmentError(
-
+        f"Unsupported graph database provider: {graph_database_provider}. "
-    graph_client = NetworkXAdapter(filename=graph_file_path)
+        f"Supported providers are: {', '.join(list(supported_databases.keys()) + ['neo4j', 'falkordb', 'kuzu', 'kuzu-remote', 'memgraph', 'neptune', 'neptune_analytics'])}"
-
+    )
-    return graph_client

cognee/infrastructure/databases/graph/neptune_driver/exceptions.py

@@ -3,11 +3,16 @@
 This module defines custom exceptions for Neptune Analytics operations.
 """
 
-from cognee.exceptions import CogneeApiError
+from cognee.exceptions import (
+    CogneeSystemError,
+    CogneeTransientError,
+    CogneeValidationError,
+    CogneeConfigurationError,
+)
 from fastapi import status
 
 
-class NeptuneAnalyticsError(CogneeApiError):
+class NeptuneAnalyticsError(CogneeSystemError):
     """Base exception for Neptune Analytics operations."""
 
     def __init__(
@@ -19,7 +24,7 @@ class NeptuneAnalyticsError(CogneeApiError):
         super().__init__(message, name, status_code)
 
 
-class NeptuneAnalyticsConnectionError(NeptuneAnalyticsError):
+class NeptuneAnalyticsConnectionError(CogneeTransientError):
     """Exception raised when connection to Neptune Analytics fails."""
 
     def __init__(
@@ -31,7 +36,7 @@ class NeptuneAnalyticsConnectionError(NeptuneAnalyticsError):
         super().__init__(message, name, status_code)
 
 
-class NeptuneAnalyticsQueryError(NeptuneAnalyticsError):
+class NeptuneAnalyticsQueryError(CogneeValidationError):
     """Exception raised when a query execution fails."""
 
     def __init__(
@@ -43,7 +48,7 @@ class NeptuneAnalyticsQueryError(NeptuneAnalyticsError):
         super().__init__(message, name, status_code)
 
 
-class NeptuneAnalyticsAuthenticationError(NeptuneAnalyticsError):
+class NeptuneAnalyticsAuthenticationError(CogneeConfigurationError):
     """Exception raised when authentication with Neptune Analytics fails."""
 
     def __init__(
@@ -55,7 +60,7 @@ class NeptuneAnalyticsAuthenticationError(NeptuneAnalyticsError):
         super().__init__(message, name, status_code)
 
 
-class NeptuneAnalyticsConfigurationError(NeptuneAnalyticsError):
+class NeptuneAnalyticsConfigurationError(CogneeConfigurationError):
     """Exception raised when Neptune Analytics configuration is invalid."""
 
     def __init__(
@@ -67,7 +72,7 @@ class NeptuneAnalyticsConfigurationError(NeptuneAnalyticsError):
         super().__init__(message, name, status_code)
 
 
-class NeptuneAnalyticsTimeoutError(NeptuneAnalyticsError):
+class NeptuneAnalyticsTimeoutError(CogneeTransientError):
     """Exception raised when a Neptune Analytics operation times out."""
 
     def __init__(
@@ -79,7 +84,7 @@ class NeptuneAnalyticsTimeoutError(NeptuneAnalyticsError):
         super().__init__(message, name, status_code)
 
 
-class NeptuneAnalyticsThrottlingError(NeptuneAnalyticsError):
+class NeptuneAnalyticsThrottlingError(CogneeTransientError):
     """Exception raised when requests are throttled by Neptune Analytics."""
 
     def __init__(
@@ -91,7 +96,7 @@ class NeptuneAnalyticsThrottlingError(NeptuneAnalyticsError):
         super().__init__(message, name, status_code)
 
 
-class NeptuneAnalyticsResourceNotFoundError(NeptuneAnalyticsError):
+class NeptuneAnalyticsResourceNotFoundError(CogneeValidationError):
     """Exception raised when a Neptune Analytics resource is not found."""
 
     def __init__(
@@ -103,7 +108,7 @@ class NeptuneAnalyticsResourceNotFoundError(NeptuneAnalyticsError):
         super().__init__(message, name, status_code)
 
 
-class NeptuneAnalyticsInvalidParameterError(NeptuneAnalyticsError):
+class NeptuneAnalyticsInvalidParameterError(CogneeValidationError):
     """Exception raised when invalid parameters are provided to Neptune Analytics."""
 
     def __init__(

cognee/infrastructure/databases/hybrid/falkordb/FalkorDBAdapter.py

@@ -9,7 +9,7 @@ from typing import List, Dict, Any, Optional, Tuple, Type, Union
 
 from falkordb import FalkorDB
 
-from cognee.exceptions import InvalidValueError
+from cognee.infrastructure.databases.exceptions import MissingQueryParameterError
 from cognee.infrastructure.databases.graph.graph_db_interface import (
     GraphDBInterface,
     record_graph_changes,
@@ -721,7 +721,7 @@ class FalkorDBAdapter(VectorDBInterface, GraphDBInterface):
             Returns the search results as a result set from the graph database.
         """
         if query_text is None and query_vector is None:
-            raise InvalidValueError(message="One of query_text or query_vector must be provided!")
+            raise MissingQueryParameterError()
 
         if query_text and not query_vector:
             query_vector = (await self.embed_data([query_text]))[0]

cognee/infrastructure/databases/hybrid/neptune_analytics/NeptuneAnalyticsAdapter.py

@@ -5,7 +5,8 @@ import json
 from typing import List, Optional, Any, Dict, Type, Tuple
 from uuid import UUID
 
-from cognee.exceptions import InvalidValueError
+from cognee.infrastructure.databases.exceptions import MissingQueryParameterError
+from cognee.infrastructure.databases.exceptions import MutuallyExclusiveQueryParametersError
 from cognee.infrastructure.databases.graph.neptune_driver.adapter import NeptuneGraphDB
 from cognee.infrastructure.databases.vector.vector_db_interface import VectorDBInterface
 from cognee.infrastructure.engine import DataPoint
@@ -274,11 +275,9 @@ class NeptuneAnalyticsAdapter(NeptuneGraphDB, VectorDBInterface):
             limit = self._TOPK_UPPER_BOUND
 
         if query_vector and query_text:
-            raise InvalidValueError(
+            raise MutuallyExclusiveQueryParametersError()
-                message="The search function accepts either text or embedding as input, but not both."
-            )
         elif query_text is None and query_vector is None:
-            raise InvalidValueError(message="One of query_text or query_vector must be provided!")
+            raise MissingQueryParameterError()
         elif query_vector:
             embedding = query_vector
         else:

cognee/infrastructure/databases/vector/chromadb/ChromaDBAdapter.py

@@ -4,13 +4,13 @@ from uuid import UUID
 from typing import List, Optional
 from chromadb import AsyncHttpClient, Settings
 
-from cognee.exceptions import InvalidValueError
 from cognee.shared.logging_utils import get_logger
 from cognee.modules.storage.utils import get_own_properties
 from cognee.infrastructure.engine import DataPoint
 from cognee.infrastructure.engine.utils import parse_id
 from cognee.infrastructure.databases.vector.exceptions import CollectionNotFoundError
 from cognee.infrastructure.databases.vector.models.ScoredResult import ScoredResult
+from cognee.infrastructure.databases.exceptions import MissingQueryParameterError
 
 from ..embeddings.EmbeddingEngine import EmbeddingEngine
 from ..vector_db_interface import VectorDBInterface
@@ -378,7 +378,7 @@ class ChromaDBAdapter(VectorDBInterface):
             Returns a list of ScoredResult instances representing the search results.
         """
         if query_text is None and query_vector is None:
-            raise InvalidValueError(message="One of query_text or query_vector must be provided!")
+            raise MissingQueryParameterError()
 
         if query_text and not query_vector:
             query_vector = (await self.embedding_engine.embed_text([query_text]))[0]

cognee/infrastructure/databases/vector/embeddings/FastembedEmbeddingEngine.py

@@ -41,11 +41,11 @@ class FastembedEmbeddingEngine(EmbeddingEngine):
         self,
         model: Optional[str] = "openai/text-embedding-3-large",
         dimensions: Optional[int] = 3072,
-        max_tokens: int = 512,
+        max_completion_tokens: int = 512,
     ):
         self.model = model
         self.dimensions = dimensions
-        self.max_tokens = max_tokens
+        self.max_completion_tokens = max_completion_tokens
         self.tokenizer = self.get_tokenizer()
         # self.retry_count = 0
         self.embedding_model = TextEmbedding(model_name=model)
@@ -112,7 +112,9 @@ class FastembedEmbeddingEngine(EmbeddingEngine):
         """
         logger.debug("Loading tokenizer for FastembedEmbeddingEngine...")
 
-        tokenizer = TikTokenTokenizer(model="gpt-4o", max_tokens=self.max_tokens)
+        tokenizer = TikTokenTokenizer(
+            model="gpt-4o", max_completion_tokens=self.max_completion_tokens
+        )
 
         logger.debug("Tokenizer loaded for for FastembedEmbeddingEngine")
         return tokenizer

cognee/infrastructure/databases/vector/embeddings/LiteLLMEmbeddingEngine.py

@@ -6,7 +6,7 @@ import math
 import litellm
 import os
 from cognee.infrastructure.databases.vector.embeddings.EmbeddingEngine import EmbeddingEngine
-from cognee.infrastructure.databases.exceptions.EmbeddingException import EmbeddingException
+from cognee.infrastructure.databases.exceptions import EmbeddingException
 from cognee.infrastructure.llm.tokenizer.Gemini import (
     GeminiTokenizer,
 )
@@ -57,7 +57,7 @@ class LiteLLMEmbeddingEngine(EmbeddingEngine):
         api_key: str = None,
         endpoint: str = None,
         api_version: str = None,
-        max_tokens: int = 512,
+        max_completion_tokens: int = 512,
     ):
         self.api_key = api_key
         self.endpoint = endpoint
@@ -65,7 +65,7 @@ class LiteLLMEmbeddingEngine(EmbeddingEngine):
         self.provider = provider
         self.model = model
         self.dimensions = dimensions
-        self.max_tokens = max_tokens
+        self.max_completion_tokens = max_completion_tokens
         self.tokenizer = self.get_tokenizer()
         self.retry_count = 0
 
@@ -179,20 +179,29 @@ class LiteLLMEmbeddingEngine(EmbeddingEngine):
         model = self.model.split("/")[-1]
 
         if "openai" in self.provider.lower():
-            tokenizer = TikTokenTokenizer(model=model, max_tokens=self.max_tokens)
+            tokenizer = TikTokenTokenizer(
+                model=model, max_completion_tokens=self.max_completion_tokens
+            )
         elif "gemini" in self.provider.lower():
-            tokenizer = GeminiTokenizer(model=model, max_tokens=self.max_tokens)
+            tokenizer = GeminiTokenizer(
+                model=model, max_completion_tokens=self.max_completion_tokens
+            )
         elif "mistral" in self.provider.lower():
-            tokenizer = MistralTokenizer(model=model, max_tokens=self.max_tokens)
+            tokenizer = MistralTokenizer(
+                model=model, max_completion_tokens=self.max_completion_tokens
+            )
         else:
             try:
                 tokenizer = HuggingFaceTokenizer(
-                    model=self.model.replace("hosted_vllm/", ""), max_tokens=self.max_tokens
+                    model=self.model.replace("hosted_vllm/", ""),
+                    max_completion_tokens=self.max_completion_tokens,
                 )
             except Exception as e:
                 logger.warning(f"Could not get tokenizer from HuggingFace due to: {e}")
                 logger.info("Switching to TikToken default tokenizer.")
-                tokenizer = TikTokenTokenizer(model=None, max_tokens=self.max_tokens)
+                tokenizer = TikTokenTokenizer(
+                    model=None, max_completion_tokens=self.max_completion_tokens
+                )
 
         logger.debug(f"Tokenizer loaded for model: {self.model}")
         return tokenizer

cognee/infrastructure/databases/vector/embeddings/OllamaEmbeddingEngine.py

@@ -30,7 +30,7 @@ class OllamaEmbeddingEngine(EmbeddingEngine):
     Instance variables:
     - model
     - dimensions
-    - max_tokens
+    - max_completion_tokens
     - endpoint
     - mock
     - huggingface_tokenizer_name
@@ -39,7 +39,7 @@ class OllamaEmbeddingEngine(EmbeddingEngine):
 
     model: str
     dimensions: int
-    max_tokens: int
+    max_completion_tokens: int
     endpoint: str
     mock: bool
     huggingface_tokenizer_name: str
@@ -50,13 +50,13 @@ class OllamaEmbeddingEngine(EmbeddingEngine):
         self,
         model: Optional[str] = "avr/sfr-embedding-mistral:latest",
         dimensions: Optional[int] = 1024,
-        max_tokens: int = 512,
+        max_completion_tokens: int = 512,
         endpoint: Optional[str] = "http://localhost:11434/api/embeddings",
         huggingface_tokenizer: str = "Salesforce/SFR-Embedding-Mistral",
     ):
         self.model = model
         self.dimensions = dimensions
-        self.max_tokens = max_tokens
+        self.max_completion_tokens = max_completion_tokens
         self.endpoint = endpoint
         self.huggingface_tokenizer_name = huggingface_tokenizer
         self.tokenizer = self.get_tokenizer()
@@ -132,7 +132,7 @@ class OllamaEmbeddingEngine(EmbeddingEngine):
         """
         logger.debug("Loading HuggingfaceTokenizer for OllamaEmbeddingEngine...")
         tokenizer = HuggingFaceTokenizer(
-            model=self.huggingface_tokenizer_name, max_tokens=self.max_tokens
+            model=self.huggingface_tokenizer_name, max_completion_tokens=self.max_completion_tokens
         )
         logger.debug("Tokenizer loaded for OllamaEmbeddingEngine")
         return tokenizer

cognee/infrastructure/databases/vector/embeddings/config.py

@@ -18,7 +18,7 @@ class EmbeddingConfig(BaseSettings):
     embedding_endpoint: Optional[str] = None
     embedding_api_key: Optional[str] = None
     embedding_api_version: Optional[str] = None
-    embedding_max_tokens: Optional[int] = 8191
+    embedding_max_completion_tokens: Optional[int] = 8191
     huggingface_tokenizer: Optional[str] = None
     model_config = SettingsConfigDict(env_file=".env", extra="allow")
 
@@ -38,7 +38,7 @@ class EmbeddingConfig(BaseSettings):
             "embedding_endpoint": self.embedding_endpoint,
             "embedding_api_key": self.embedding_api_key,
             "embedding_api_version": self.embedding_api_version,
-            "embedding_max_tokens": self.embedding_max_tokens,
+            "embedding_max_completion_tokens": self.embedding_max_completion_tokens,
             "huggingface_tokenizer": self.huggingface_tokenizer,
         }
 

cognee/infrastructure/databases/vector/embeddings/get_embedding_engine.py

@@ -27,7 +27,7 @@ def get_embedding_engine() -> EmbeddingEngine:
         config.embedding_provider,
         config.embedding_model,
         config.embedding_dimensions,
-        config.embedding_max_tokens,
+        config.embedding_max_completion_tokens,
         config.embedding_endpoint,
         config.embedding_api_key,
         config.embedding_api_version,
@@ -41,7 +41,7 @@ def create_embedding_engine(
     embedding_provider,
     embedding_model,
     embedding_dimensions,
-    embedding_max_tokens,
+    embedding_max_completion_tokens,
     embedding_endpoint,
     embedding_api_key,
     embedding_api_version,
@@ -58,7 +58,7 @@ def create_embedding_engine(
           'ollama', or another supported provider.
         - embedding_model: The model to be used for the embedding engine.
         - embedding_dimensions: The number of dimensions for the embeddings.
-        - embedding_max_tokens: The maximum number of tokens for the embeddings.
+        - embedding_max_completion_tokens: The maximum number of tokens for the embeddings.
         - embedding_endpoint: The endpoint for the embedding service, relevant for certain
           providers.
         - embedding_api_key: API key to authenticate with the embedding service, if
@@ -81,7 +81,7 @@ def create_embedding_engine(
         return FastembedEmbeddingEngine(
             model=embedding_model,
             dimensions=embedding_dimensions,
-            max_tokens=embedding_max_tokens,
+            max_completion_tokens=embedding_max_completion_tokens,
         )
 
     if embedding_provider == "ollama":
@@ -90,7 +90,7 @@ def create_embedding_engine(
         return OllamaEmbeddingEngine(
             model=embedding_model,
             dimensions=embedding_dimensions,
-            max_tokens=embedding_max_tokens,
+            max_completion_tokens=embedding_max_completion_tokens,
             endpoint=embedding_endpoint,
             huggingface_tokenizer=huggingface_tokenizer,
         )
@@ -104,5 +104,5 @@ def create_embedding_engine(
         api_version=embedding_api_version,
         model=embedding_model,
         dimensions=embedding_dimensions,
-        max_tokens=embedding_max_tokens,
+        max_completion_tokens=embedding_max_completion_tokens,
     )

cognee/infrastructure/databases/vector/exceptions/exceptions.py

@@ -1,12 +1,12 @@
 from fastapi import status
-from cognee.exceptions import CriticalError
+from cognee.exceptions import CogneeValidationError
 
 
-class CollectionNotFoundError(CriticalError):
+class CollectionNotFoundError(CogneeValidationError):
     """
     Represents an error that occurs when a requested collection cannot be found.
 
-    This class extends the CriticalError to handle specific cases where a requested
+    This class extends the CogneeValidationError to handle specific cases where a requested
     collection is unavailable. It can be initialized with a custom message and allows for
     logging options including log level and whether to log the error.
     """

cognee/infrastructure/databases/vector/lancedb/LanceDBAdapter.py

@@ -5,7 +5,7 @@ from pydantic import BaseModel
 from lancedb.pydantic import LanceModel, Vector
 from typing import Generic, List, Optional, TypeVar, Union, get_args, get_origin, get_type_hints
 
-from cognee.exceptions import InvalidValueError
+from cognee.infrastructure.databases.exceptions import MissingQueryParameterError
 from cognee.infrastructure.engine import DataPoint
 from cognee.infrastructure.engine.utils import parse_id
 from cognee.infrastructure.files.storage import get_file_storage
@@ -228,7 +228,7 @@ class LanceDBAdapter(VectorDBInterface):
         normalized: bool = True,
     ):
         if query_text is None and query_vector is None:
-            raise InvalidValueError(message="One of query_text or query_vector must be provided!")
+            raise MissingQueryParameterError()
 
         if query_text and not query_vector:
             query_vector = (await self.embedding_engine.embed_text([query_text]))[0]

cognee/infrastructure/databases/vector/pgvector/PGVectorAdapter.py

@@ -9,7 +9,7 @@ from sqlalchemy.exc import ProgrammingError
 from tenacity import retry, retry_if_exception_type, stop_after_attempt, wait_exponential
 from asyncpg import DeadlockDetectedError, DuplicateTableError, UniqueViolationError
 
-from cognee.exceptions import InvalidValueError
+
 from cognee.shared.logging_utils import get_logger
 from cognee.infrastructure.engine import DataPoint
 from cognee.infrastructure.engine.utils import parse_id
@@ -17,6 +17,7 @@ from cognee.infrastructure.databases.relational import get_relational_engine
 
 from distributed.utils import override_distributed
 from distributed.tasks.queued_add_data_points import queued_add_data_points
+from cognee.infrastructure.databases.exceptions import MissingQueryParameterError
 
 from ...relational.ModelBase import Base
 from ...relational.sqlalchemy.SqlAlchemyAdapter import SQLAlchemyAdapter
@@ -275,7 +276,7 @@ class PGVectorAdapter(SQLAlchemyAdapter, VectorDBInterface):
                 return metadata.tables[collection_name]
             else:
                 raise CollectionNotFoundError(
-                    f"Collection '{collection_name}' not found!", log_level="DEBUG"
+                    f"Collection '{collection_name}' not found!",
                 )
 
     async def retrieve(self, collection_name: str, data_point_ids: List[str]):
@@ -302,7 +303,7 @@ class PGVectorAdapter(SQLAlchemyAdapter, VectorDBInterface):
         with_vector: bool = False,
     ) -> List[ScoredResult]:
         if query_text is None and query_vector is None:
-            raise InvalidValueError(message="One of query_text or query_vector must be provided!")
+            raise MissingQueryParameterError()
 
         if query_text and not query_vector:
             query_vector = (await self.embedding_engine.embed_text([query_text]))[0]

cognee/infrastructure/llm/config.py

@@ -18,7 +18,7 @@ class LLMConfig(BaseSettings):
     - llm_api_version
     - llm_temperature
     - llm_streaming
-    - llm_max_tokens
+    - llm_max_completion_tokens
     - transcription_model
     - graph_prompt_path
     - llm_rate_limit_enabled
@@ -35,16 +35,16 @@ class LLMConfig(BaseSettings):
 
     structured_output_framework: str = "instructor"
     llm_provider: str = "openai"
-    llm_model: str = "gpt-4o-mini"
+    llm_model: str = "gpt-5-mini"
     llm_endpoint: str = ""
     llm_api_key: Optional[str] = None
     llm_api_version: Optional[str] = None
     llm_temperature: float = 0.0
     llm_streaming: bool = False
-    llm_max_tokens: int = 16384
+    llm_max_completion_tokens: int = 16384
 
     baml_llm_provider: str = "openai"
-    baml_llm_model: str = "gpt-4o-mini"
+    baml_llm_model: str = "gpt-5-mini"
     baml_llm_endpoint: str = ""
     baml_llm_api_key: Optional[str] = None
     baml_llm_temperature: float = 0.0
@@ -171,7 +171,7 @@ class LLMConfig(BaseSettings):
             "api_version": self.llm_api_version,
             "temperature": self.llm_temperature,
             "streaming": self.llm_streaming,
-            "max_tokens": self.llm_max_tokens,
+            "max_completion_tokens": self.llm_max_completion_tokens,
             "transcription_model": self.transcription_model,
             "graph_prompt_path": self.graph_prompt_path,
             "rate_limit_enabled": self.llm_rate_limit_enabled,

cognee/infrastructure/llm/exceptions.py

@@ -1,5 +1,33 @@
-from cognee.exceptions.exceptions import CriticalError
+from cognee.exceptions.exceptions import CogneeValidationError
 
 
-class ContentPolicyFilterError(CriticalError):
+class ContentPolicyFilterError(CogneeValidationError):
     pass
+
+
+class LLMAPIKeyNotSetError(CogneeValidationError):
+    """
+    Raised when the LLM API key is not set in the configuration.
+    """
+
+    def __init__(self, message: str = "LLM API key is not set."):
+        super().__init__(message=message, name="LLMAPIKeyNotSetError")
+
+
+class UnsupportedLLMProviderError(CogneeValidationError):
+    """
+    Raised when an unsupported LLM provider is specified in the configuration.
+    """
+
+    def __init__(self, provider: str):
+        message = f"Unsupported LLM provider: {provider}"
+        super().__init__(message=message, name="UnsupportedLLMProviderError")
+
+
+class MissingSystemPromptPathError(CogneeValidationError):
+    def __init__(
+        self,
+        name: str = "MissingSystemPromptPathError",
+    ):
+        message = "No system prompt path provided."
+        super().__init__(message, name)

cognee/infrastructure/llm/structured_output_framework/litellm_instructor/llm/anthropic/adapter.py

@@ -2,7 +2,7 @@ from typing import Type
 from pydantic import BaseModel
 import instructor
 
-from cognee.exceptions import InvalidValueError
+from cognee.infrastructure.llm.exceptions import MissingSystemPromptPathError
 from cognee.infrastructure.llm.structured_output_framework.litellm_instructor.llm.llm_interface import (
     LLMInterface,
 )
@@ -23,7 +23,7 @@ class AnthropicAdapter(LLMInterface):
     name = "Anthropic"
     model: str
 
-    def __init__(self, max_tokens: int, model: str = None):
+    def __init__(self, max_completion_tokens: int, model: str = None):
         import anthropic
 
         self.aclient = instructor.patch(
@@ -31,7 +31,7 @@ class AnthropicAdapter(LLMInterface):
         )
 
         self.model = model
-        self.max_tokens = max_tokens
+        self.max_completion_tokens = max_completion_tokens
 
     @sleep_and_retry_async()
     @rate_limit_async
@@ -57,7 +57,7 @@ class AnthropicAdapter(LLMInterface):
 
         return await self.aclient(
             model=self.model,
-            max_tokens=4096,
+            max_completion_tokens=4096,
             max_retries=5,
             messages=[
                 {
@@ -89,7 +89,7 @@ class AnthropicAdapter(LLMInterface):
         if not text_input:
             text_input = "No user input provided."
         if not system_prompt:
-            raise InvalidValueError(message="No system prompt path provided.")
+            raise MissingSystemPromptPathError()
 
         system_prompt = LLMGateway.read_query_prompt(system_prompt)
 

cognee/infrastructure/llm/structured_output_framework/litellm_instructor/llm/gemini/adapter.py

@@ -5,7 +5,7 @@ from litellm import acompletion, JSONSchemaValidationError
 
 from cognee.shared.logging_utils import get_logger
 from cognee.modules.observability.get_observe import get_observe
-from cognee.exceptions import InvalidValueError
+from cognee.infrastructure.llm.exceptions import MissingSystemPromptPathError
 from cognee.infrastructure.llm.structured_output_framework.litellm_instructor.llm.llm_interface import (
     LLMInterface,
 )
@@ -34,7 +34,7 @@ class GeminiAdapter(LLMInterface):
         self,
         api_key: str,
         model: str,
-        max_tokens: int,
+        max_completion_tokens: int,
         endpoint: Optional[str] = None,
         api_version: Optional[str] = None,
         streaming: bool = False,
@@ -44,7 +44,7 @@ class GeminiAdapter(LLMInterface):
         self.endpoint = endpoint
         self.api_version = api_version
         self.streaming = streaming
-        self.max_tokens = max_tokens
+        self.max_completion_tokens = max_completion_tokens
 
     @observe(as_type="generation")
     @sleep_and_retry_async()
@@ -90,7 +90,7 @@ class GeminiAdapter(LLMInterface):
                     model=f"{self.model}",
                     messages=messages,
                     api_key=self.api_key,
-                    max_tokens=self.max_tokens,
+                    max_completion_tokens=self.max_completion_tokens,
                     temperature=0.1,
                     response_format=response_schema,
                     timeout=100,
@@ -118,7 +118,7 @@ class GeminiAdapter(LLMInterface):
         """
         Format and display the prompt for a user query.
 
-        Raises an InvalidValueError if no system prompt is provided.
+        Raises an MissingQueryParameterError if no system prompt is provided.
 
         Parameters:
         -----------
@@ -135,7 +135,7 @@ class GeminiAdapter(LLMInterface):
         if not text_input:
             text_input = "No user input provided."
         if not system_prompt:
-            raise InvalidValueError(message="No system prompt path provided.")
+            raise MissingSystemPromptPathError()
         system_prompt = LLMGateway.read_query_prompt(system_prompt)
 
         formatted_prompt = (

cognee/infrastructure/llm/structured_output_framework/litellm_instructor/llm/generic_llm_api/adapter.py

@@ -41,7 +41,7 @@ class GenericAPIAdapter(LLMInterface):
         api_key: str,
         model: str,
         name: str,
-        max_tokens: int,
+        max_completion_tokens: int,
         fallback_model: str = None,
         fallback_api_key: str = None,
         fallback_endpoint: str = None,
@@ -50,7 +50,7 @@ class GenericAPIAdapter(LLMInterface):
         self.model = model
         self.api_key = api_key
         self.endpoint = endpoint
-        self.max_tokens = max_tokens
+        self.max_completion_tokens = max_completion_tokens
 
         self.fallback_model = fallback_model
         self.fallback_api_key = fallback_api_key

cognee/infrastructure/llm/structured_output_framework/litellm_instructor/llm/get_llm_client.py

@@ -2,11 +2,14 @@
 
 from enum import Enum
 
-from cognee.exceptions import InvalidValueError
 from cognee.infrastructure.llm import get_llm_config
 from cognee.infrastructure.llm.structured_output_framework.litellm_instructor.llm.ollama.adapter import (
     OllamaAPIAdapter,
 )
+from cognee.infrastructure.llm.exceptions import (
+    LLMAPIKeyNotSetError,
+    UnsupportedLLMProviderError,
+)
 
 
 # Define an Enum for LLM Providers
@@ -35,7 +38,7 @@ def get_llm_client():
 
     This function retrieves the configuration for the LLM provider and model, and
     initializes the appropriate LLM client adapter accordingly. It raises an
-    InvalidValueError if the LLM API key is not set for certain providers or if the provider
+    LLMAPIKeyNotSetError if the LLM API key is not set for certain providers or if the provider
     is unsupported.
 
     Returns:
@@ -51,15 +54,19 @@ def get_llm_client():
     # Check if max_token value is defined in liteLLM for given model
     # if not use value from cognee configuration
     from cognee.infrastructure.llm.utils import (
-        get_model_max_tokens,
+        get_model_max_completion_tokens,
     )  # imported here to avoid circular imports
 
-    model_max_tokens = get_model_max_tokens(llm_config.llm_model)
+    model_max_completion_tokens = get_model_max_completion_tokens(llm_config.llm_model)
-    max_tokens = model_max_tokens if model_max_tokens else llm_config.llm_max_tokens
+    max_completion_tokens = (
+        model_max_completion_tokens
+        if model_max_completion_tokens
+        else llm_config.llm_max_completion_tokens
+    )
 
     if provider == LLMProvider.OPENAI:
         if llm_config.llm_api_key is None:
-            raise InvalidValueError(message="LLM API key is not set.")
+            raise LLMAPIKeyNotSetError()
 
         from cognee.infrastructure.llm.structured_output_framework.litellm_instructor.llm.openai.adapter import (
             OpenAIAdapter,
@@ -71,7 +78,7 @@ def get_llm_client():
             api_version=llm_config.llm_api_version,
             model=llm_config.llm_model,
             transcription_model=llm_config.transcription_model,
-            max_tokens=max_tokens,
+            max_completion_tokens=max_completion_tokens,
             streaming=llm_config.llm_streaming,
             fallback_api_key=llm_config.fallback_api_key,
             fallback_endpoint=llm_config.fallback_endpoint,
@@ -80,7 +87,7 @@ def get_llm_client():
 
     elif provider == LLMProvider.OLLAMA:
         if llm_config.llm_api_key is None:
-            raise InvalidValueError(message="LLM API key is not set.")
+            raise LLMAPIKeyNotSetError()
 
         from cognee.infrastructure.llm.structured_output_framework.litellm_instructor.llm.generic_llm_api.adapter import (
             GenericAPIAdapter,
@@ -91,7 +98,7 @@ def get_llm_client():
             llm_config.llm_api_key,
             llm_config.llm_model,
             "Ollama",
-            max_tokens=max_tokens,
+            max_completion_tokens=max_completion_tokens,
         )
 
     elif provider == LLMProvider.ANTHROPIC:
@@ -99,11 +106,13 @@ def get_llm_client():
             AnthropicAdapter,
         )
 
-        return AnthropicAdapter(max_tokens=max_tokens, model=llm_config.llm_model)
+        return AnthropicAdapter(
+            max_completion_tokens=max_completion_tokens, model=llm_config.llm_model
+        )
 
     elif provider == LLMProvider.CUSTOM:
         if llm_config.llm_api_key is None:
-            raise InvalidValueError(message="LLM API key is not set.")
+            raise LLMAPIKeyNotSetError()
 
         from cognee.infrastructure.llm.structured_output_framework.litellm_instructor.llm.generic_llm_api.adapter import (
             GenericAPIAdapter,
@@ -114,7 +123,7 @@ def get_llm_client():
             llm_config.llm_api_key,
             llm_config.llm_model,
             "Custom",
-            max_tokens=max_tokens,
+            max_completion_tokens=max_completion_tokens,
             fallback_api_key=llm_config.fallback_api_key,
             fallback_endpoint=llm_config.fallback_endpoint,
             fallback_model=llm_config.fallback_model,
@@ -122,7 +131,7 @@ def get_llm_client():
 
     elif provider == LLMProvider.GEMINI:
         if llm_config.llm_api_key is None:
-            raise InvalidValueError(message="LLM API key is not set.")
+            raise LLMAPIKeyNotSetError()
 
         from cognee.infrastructure.llm.structured_output_framework.litellm_instructor.llm.gemini.adapter import (
             GeminiAdapter,
@@ -131,11 +140,11 @@ def get_llm_client():
         return GeminiAdapter(
             api_key=llm_config.llm_api_key,
             model=llm_config.llm_model,
-            max_tokens=max_tokens,
+            max_completion_tokens=max_completion_tokens,
             endpoint=llm_config.llm_endpoint,
             api_version=llm_config.llm_api_version,
             streaming=llm_config.llm_streaming,
         )
 
     else:
-        raise InvalidValueError(message=f"Unsupported LLM provider: {provider}")
+        raise UnsupportedLLMProviderError(provider)

cognee/infrastructure/llm/structured_output_framework/litellm_instructor/llm/ollama/adapter.py

@@ -30,16 +30,18 @@ class OllamaAPIAdapter(LLMInterface):
     - model
     - api_key
     - endpoint
-    - max_tokens
+    - max_completion_tokens
     - aclient
     """
 
-    def __init__(self, endpoint: str, api_key: str, model: str, name: str, max_tokens: int):
+    def __init__(
+        self, endpoint: str, api_key: str, model: str, name: str, max_completion_tokens: int
+    ):
         self.name = name
         self.model = model
         self.api_key = api_key
         self.endpoint = endpoint
-        self.max_tokens = max_tokens
+        self.max_completion_tokens = max_completion_tokens
 
         self.aclient = instructor.from_openai(
             OpenAI(base_url=self.endpoint, api_key=self.api_key), mode=instructor.Mode.JSON
@@ -159,7 +161,7 @@ class OllamaAPIAdapter(LLMInterface):
                     ],
                 }
             ],
-            max_tokens=300,
+            max_completion_tokens=300,
         )
 
         # Ensure response is valid before accessing .choices[0].message.content

cognee/infrastructure/llm/structured_output_framework/litellm_instructor/llm/openai/adapter.py

@@ -7,12 +7,14 @@ from openai import ContentFilterFinishReasonError
 from litellm.exceptions import ContentPolicyViolationError
 from instructor.exceptions import InstructorRetryException
 
-from cognee.exceptions import InvalidValueError
 from cognee.infrastructure.llm.LLMGateway import LLMGateway
 from cognee.infrastructure.llm.structured_output_framework.litellm_instructor.llm.llm_interface import (
     LLMInterface,
 )
-from cognee.infrastructure.llm.exceptions import ContentPolicyFilterError
+from cognee.infrastructure.llm.exceptions import (
+    ContentPolicyFilterError,
+    MissingSystemPromptPathError,
+)
 from cognee.infrastructure.files.utils.open_data_file import open_data_file
 from cognee.infrastructure.llm.structured_output_framework.litellm_instructor.llm.rate_limiter import (
     rate_limit_async,
@@ -62,7 +64,7 @@ class OpenAIAdapter(LLMInterface):
         api_version: str,
         model: str,
         transcription_model: str,
-        max_tokens: int,
+        max_completion_tokens: int,
         streaming: bool = False,
         fallback_model: str = None,
         fallback_api_key: str = None,
@@ -75,7 +77,7 @@ class OpenAIAdapter(LLMInterface):
         self.api_key = api_key
         self.endpoint = endpoint
         self.api_version = api_version
-        self.max_tokens = max_tokens
+        self.max_completion_tokens = max_completion_tokens
         self.streaming = streaming
 
         self.fallback_model = fallback_model
@@ -299,7 +301,7 @@ class OpenAIAdapter(LLMInterface):
             api_key=self.api_key,
             api_base=self.endpoint,
             api_version=self.api_version,
-            max_tokens=300,
+            max_completion_tokens=300,
             max_retries=self.MAX_RETRIES,
         )
 
@@ -308,7 +310,7 @@ class OpenAIAdapter(LLMInterface):
         Format and display the prompt for a user query.
 
         This method formats the prompt using the provided user input and system prompt,
-        returning a string representation. Raises InvalidValueError if the system prompt is not
+        returning a string representation. Raises MissingSystemPromptPathError if the system prompt is not
         provided.
 
         Parameters:
@@ -325,7 +327,7 @@ class OpenAIAdapter(LLMInterface):
         if not text_input:
             text_input = "No user input provided."
         if not system_prompt:
-            raise InvalidValueError(message="No system prompt path provided.")
+            raise MissingSystemPromptPathError()
         system_prompt = LLMGateway.read_query_prompt(system_prompt)
 
         formatted_prompt = (

cognee/infrastructure/llm/tokenizer/Gemini/adapter.py

@@ -17,10 +17,10 @@ class GeminiTokenizer(TokenizerInterface):
     def __init__(
         self,
         model: str,
-        max_tokens: int = 3072,
+        max_completion_tokens: int = 3072,
     ):
         self.model = model
-        self.max_tokens = max_tokens
+        self.max_completion_tokens = max_completion_tokens
 
         # Get LLM API key from config
         from cognee.infrastructure.databases.vector.embeddings.config import get_embedding_config

cognee/infrastructure/llm/tokenizer/HuggingFace/adapter.py

@@ -14,17 +14,17 @@ class HuggingFaceTokenizer(TokenizerInterface):
 
     Instance variables include:
     - model: str
-    - max_tokens: int
+    - max_completion_tokens: int
     - tokenizer: AutoTokenizer
     """
 
     def __init__(
         self,
         model: str,
-        max_tokens: int = 512,
+        max_completion_tokens: int = 512,
     ):
         self.model = model
-        self.max_tokens = max_tokens
+        self.max_completion_tokens = max_completion_tokens
 
         # Import here to make it an optional dependency
         from transformers import AutoTokenizer

cognee/infrastructure/llm/tokenizer/Mistral/adapter.py

@@ -16,17 +16,17 @@ class MistralTokenizer(TokenizerInterface):
 
     Instance variables include:
     - model: str
-    - max_tokens: int
+    - max_completion_tokens: int
     - tokenizer: MistralTokenizer
     """
 
     def __init__(
         self,
         model: str,
-        max_tokens: int = 3072,
+        max_completion_tokens: int = 3072,
     ):
         self.model = model
-        self.max_tokens = max_tokens
+        self.max_completion_tokens = max_completion_tokens
 
         # Import here to make it an optional dependency
         from mistral_common.tokens.tokenizers.mistral import MistralTokenizer

cognee/infrastructure/llm/tokenizer/TikToken/adapter.py

@@ -13,10 +13,10 @@ class TikTokenTokenizer(TokenizerInterface):
     def __init__(
         self,
         model: Optional[str] = None,
-        max_tokens: int = 8191,
+        max_completion_tokens: int = 8191,
     ):
         self.model = model
-        self.max_tokens = max_tokens
+        self.max_completion_tokens = max_completion_tokens
         # Initialize TikToken for GPT based on model
         if model:
             self.tokenizer = tiktoken.encoding_for_model(self.model)
@@ -93,9 +93,9 @@ class TikTokenTokenizer(TokenizerInterface):
         num_tokens = len(self.tokenizer.encode(text))
         return num_tokens
 
-    def trim_text_to_max_tokens(self, text: str) -> str:
+    def trim_text_to_max_completion_tokens(self, text: str) -> str:
         """
-        Trim the text so that the number of tokens does not exceed max_tokens.
+        Trim the text so that the number of tokens does not exceed max_completion_tokens.
 
         Parameters:
         -----------
@@ -111,13 +111,13 @@ class TikTokenTokenizer(TokenizerInterface):
         num_tokens = self.count_tokens(text)
 
         # If the number of tokens is within the limit, return the text as is
-        if num_tokens <= self.max_tokens:
+        if num_tokens <= self.max_completion_tokens:
             return text
 
         # If the number exceeds the limit, trim the text
         # This is a simple trim, it may cut words in half; consider using word boundaries for a cleaner cut
         encoded_text = self.tokenizer.encode(text)
-        trimmed_encoded_text = encoded_text[: self.max_tokens]
+        trimmed_encoded_text = encoded_text[: self.max_completion_tokens]
         # Decoding the trimmed text
         trimmed_text = self.tokenizer.decode(trimmed_encoded_text)
         return trimmed_text

cognee/infrastructure/llm/utils.py

@@ -32,13 +32,13 @@ def get_max_chunk_tokens():
 
     # We need to make sure chunk size won't take more than half of LLM max context token size
     # but it also can't be bigger than the embedding engine max token size
-    llm_cutoff_point = llm_client.max_tokens // 2  # Round down the division
+    llm_cutoff_point = llm_client.max_completion_tokens // 2  # Round down the division
-    max_chunk_tokens = min(embedding_engine.max_tokens, llm_cutoff_point)
+    max_chunk_tokens = min(embedding_engine.max_completion_tokens, llm_cutoff_point)
 
     return max_chunk_tokens
 
 
-def get_model_max_tokens(model_name: str):
+def get_model_max_completion_tokens(model_name: str):
     """
     Retrieve the maximum token limit for a specified model name if it exists.
 
@@ -56,15 +56,15 @@ def get_model_max_tokens(model_name: str):
 
         Number of max tokens of model, or None if model is unknown
     """
-    max_tokens = None
+    max_completion_tokens = None
 
     if model_name in litellm.model_cost:
-        max_tokens = litellm.model_cost[model_name]["max_tokens"]
+        max_completion_tokens = litellm.model_cost[model_name]["max_tokens"]
-        logger.debug(f"Max input tokens for {model_name}: {max_tokens}")
+        logger.debug(f"Max input tokens for {model_name}: {max_completion_tokens}")
     else:
         logger.info("Model not found in LiteLLM's model_cost.")
 
-    return max_tokens
+    return max_completion_tokens
 
 
 async def test_llm_connection():

cognee/infrastructure/loaders/LoaderInterface.py

@@ -58,7 +58,7 @@ class LoaderInterface(ABC):
         pass
 
     @abstractmethod
-    async def load(self, file_path: str, file_stream: Optional[Any] = None, **kwargs):
+    async def load(self, file_path: str, **kwargs):
         """
         Load and process the file, returning standardized result.
 

cognee/modules/data/exceptions/exceptions.py

@@ -1,8 +1,11 @@
-from cognee.exceptions import CogneeApiError
+from cognee.exceptions import (
+    CogneeValidationError,
+    CogneeConfigurationError,
+)
 from fastapi import status
 
 
-class UnstructuredLibraryImportError(CogneeApiError):
+class UnstructuredLibraryImportError(CogneeConfigurationError):
     def __init__(
         self,
         message: str = "Import error. Unstructured library is not installed.",
@@ -12,7 +15,7 @@ class UnstructuredLibraryImportError(CogneeApiError):
         super().__init__(message, name, status_code)
 
 
-class UnauthorizedDataAccessError(CogneeApiError):
+class UnauthorizedDataAccessError(CogneeValidationError):
     def __init__(
         self,
         message: str = "User does not have permission to access this data.",
@@ -22,7 +25,7 @@ class UnauthorizedDataAccessError(CogneeApiError):
         super().__init__(message, name, status_code)
 
 
-class DatasetNotFoundError(CogneeApiError):
+class DatasetNotFoundError(CogneeValidationError):
     def __init__(
         self,
         message: str = "Dataset not found.",
@@ -32,7 +35,7 @@ class DatasetNotFoundError(CogneeApiError):
         super().__init__(message, name, status_code)
 
 
-class DatasetTypeError(CogneeApiError):
+class DatasetTypeError(CogneeValidationError):
     def __init__(
         self,
         message: str = "Dataset type not supported.",
@@ -40,3 +43,13 @@ class DatasetTypeError(CogneeApiError):
         status_code=status.HTTP_400_BAD_REQUEST,
     ):
         super().__init__(message, name, status_code)
+
+
+class InvalidTableAttributeError(CogneeValidationError):
+    def __init__(
+        self,
+        message: str = "The provided data object is missing the required '__tablename__' attribute.",
+        name: str = "InvalidTableAttributeError",
+        status_code: int = status.HTTP_400_BAD_REQUEST,
+    ):
+        super().__init__(message, name, status_code)

cognee/modules/data/methods/delete_data.py

@@ -1,4 +1,4 @@
-from cognee.exceptions import InvalidAttributeError
+from cognee.modules.data.exceptions.exceptions import InvalidTableAttributeError
 from cognee.modules.data.models import Data
 from cognee.infrastructure.databases.relational import get_relational_engine
 
@@ -13,9 +13,7 @@ async def delete_data(data: Data):
         ValueError: If the data object is invalid.
     """
     if not hasattr(data, "__tablename__"):
-        raise InvalidAttributeError(
+        raise InvalidTableAttributeError()
-            message="The provided data object is missing the required '__tablename__' attribute."
-        )
 
     db_engine = get_relational_engine()
 

cognee/modules/data/processing/document_types/exceptions/exceptions.py

@@ -1,8 +1,8 @@
-from cognee.exceptions import CogneeApiError
+from cognee.exceptions import CogneeSystemError
 from fastapi import status
 
 
-class PyPdfInternalError(CogneeApiError):
+class PyPdfInternalError(CogneeSystemError):
     """Internal pypdf error"""
 
     def __init__(

cognee/modules/graph/cognee_graph/CogneeGraph.py

@@ -2,8 +2,11 @@ import time
 from cognee.shared.logging_utils import get_logger
 from typing import List, Dict, Union, Optional, Type
 
-from cognee.exceptions import InvalidValueError
+from cognee.modules.graph.exceptions import (
-from cognee.modules.graph.exceptions import EntityNotFoundError, EntityAlreadyExistsError
+    EntityNotFoundError,
+    EntityAlreadyExistsError,
+    InvalidDimensionsError,
+)
 from cognee.infrastructure.databases.graph.graph_db_interface import GraphDBInterface
 from cognee.modules.graph.cognee_graph.CogneeGraphElements import Node, Edge
 from cognee.modules.graph.cognee_graph.CogneeAbstractGraph import CogneeAbstractGraph
@@ -66,8 +69,7 @@ class CogneeGraph(CogneeAbstractGraph):
         node_name: Optional[List[str]] = None,
     ) -> None:
         if node_dimension < 1 or edge_dimension < 1:
-            raise InvalidValueError(message="Dimensions must be positive integers")
+            raise InvalidDimensionsError()
-
         try:
             import time
 

cognee/modules/graph/cognee_graph/CogneeGraphElements.py

@@ -1,7 +1,6 @@
 import numpy as np
 from typing import List, Dict, Optional, Any, Union
-
+from cognee.modules.graph.exceptions import InvalidDimensionsError, DimensionOutOfRangeError
-from cognee.exceptions import InvalidValueError
 
 
 class Node:
@@ -24,7 +23,7 @@ class Node:
         self, node_id: str, attributes: Optional[Dict[str, Any]] = None, dimension: int = 1
     ):
         if dimension <= 0:
-            raise InvalidValueError(message="Dimension must be a positive integer")
+            raise InvalidDimensionsError()
         self.id = node_id
         self.attributes = attributes if attributes is not None else {}
         self.attributes["vector_distance"] = float("inf")
@@ -58,9 +57,7 @@ class Node:
 
     def is_node_alive_in_dimension(self, dimension: int) -> bool:
         if dimension < 0 or dimension >= len(self.status):
-            raise InvalidValueError(
+            raise DimensionOutOfRangeError(dimension=dimension, max_index=len(self.status) - 1)
-                message=f"Dimension {dimension} is out of range. Valid range is 0 to {len(self.status) - 1}."
-            )
         return self.status[dimension] == 1
 
     def add_attribute(self, key: str, value: Any) -> None:
@@ -110,7 +107,7 @@ class Edge:
         dimension: int = 1,
     ):
         if dimension <= 0:
-            raise InvalidValueError(message="Dimensions must be a positive integer.")
+            raise InvalidDimensionsError()
         self.node1 = node1
         self.node2 = node2
         self.attributes = attributes if attributes is not None else {}
@@ -120,9 +117,7 @@ class Edge:
 
     def is_edge_alive_in_dimension(self, dimension: int) -> bool:
         if dimension < 0 or dimension >= len(self.status):
-            raise InvalidValueError(
+            raise DimensionOutOfRangeError(dimension=dimension, max_index=len(self.status) - 1)
-                message=f"Dimension {dimension} is out of range. Valid range is 0 to {len(self.status) - 1}."
-            )
         return self.status[dimension] == 1
 
     def add_attribute(self, key: str, value: Any) -> None:

cognee/modules/graph/exceptions/__init__.py

@@ -7,4 +7,6 @@ This module defines a set of exceptions for handling various graph errors
 from .exceptions import (
     EntityNotFoundError,
     EntityAlreadyExistsError,
+    InvalidDimensionsError,
+    DimensionOutOfRangeError,
 )

cognee/modules/graph/exceptions/exceptions.py

@@ -1,8 +1,8 @@
-from cognee.exceptions import CogneeApiError
+from cognee.exceptions import CogneeValidationError
 from fastapi import status
 
 
-class EntityNotFoundError(CogneeApiError):
+class EntityNotFoundError(CogneeValidationError):
     """Database returns nothing"""
 
     def __init__(
@@ -14,7 +14,7 @@ class EntityNotFoundError(CogneeApiError):
         super().__init__(message, name, status_code)
 
 
-class EntityAlreadyExistsError(CogneeApiError):
+class EntityAlreadyExistsError(CogneeValidationError):
     """Conflict detected, like trying to create a resource that already exists"""
 
     def __init__(
@@ -24,3 +24,25 @@ class EntityAlreadyExistsError(CogneeApiError):
         status_code=status.HTTP_409_CONFLICT,
     ):
         super().__init__(message, name, status_code)
+
+
+class InvalidDimensionsError(CogneeValidationError):
+    def __init__(
+        self,
+        name: str = "InvalidDimensionsError",
+        status_code: int = status.HTTP_400_BAD_REQUEST,
+    ):
+        message = "Dimensions must be positive integers."
+        super().__init__(message, name, status_code)
+
+
+class DimensionOutOfRangeError(CogneeValidationError):
+    def __init__(
+        self,
+        dimension: int,
+        max_index: int,
+        name: str = "DimensionOutOfRangeError",
+        status_code: int = status.HTTP_400_BAD_REQUEST,
+    ):
+        message = f"Dimension {dimension} is out of range. Valid range is 0 to {max_index}."
+        super().__init__(message, name, status_code)

cognee/modules/ingestion/exceptions/exceptions.py

@@ -1,8 +1,8 @@
-from cognee.exceptions import CogneeApiError
+from cognee.exceptions import CogneeValidationError
 from fastapi import status
 
 
-class IngestionError(CogneeApiError):
+class IngestionError(CogneeValidationError):
     def __init__(
         self,
         message: str = "Type of data sent to classify not supported.",

cognee/modules/ontology/exceptions/exceptions.py

@@ -1,8 +1,8 @@
-from cognee.exceptions import CogneeApiError
+from cognee.exceptions import CogneeSystemError
 from fastapi import status
 
 
-class OntologyInitializationError(CogneeApiError):
+class OntologyInitializationError(CogneeSystemError):
     def __init__(
         self,
         message: str = "Ontology initialization failed",
@@ -12,7 +12,7 @@ class OntologyInitializationError(CogneeApiError):
         super().__init__(message, name, status_code)
 
 
-class FindClosestMatchError(CogneeApiError):
+class FindClosestMatchError(CogneeSystemError):
     def __init__(
         self,
         message: str = "Error in find_closest_match",
@@ -22,7 +22,7 @@ class FindClosestMatchError(CogneeApiError):
         super().__init__(message, name, status_code)
 
 
-class GetSubgraphError(CogneeApiError):
+class GetSubgraphError(CogneeSystemError):
     def __init__(
         self,
         message: str = "Failed to retrieve subgraph",

cognee/modules/pipelines/exceptions/exceptions.py

@@ -1,8 +1,8 @@
-from cognee.exceptions import CogneeApiError
+from cognee.exceptions import CogneeSystemError
 from fastapi import status
 
 
-class PipelineRunFailedError(CogneeApiError):
+class PipelineRunFailedError(CogneeSystemError):
     def __init__(
         self,
         message: str = "Pipeline run failed.",

cognee/modules/retrieval/cypher_search_retriever.py

@@ -1,6 +1,5 @@
 from typing import Any, Optional
 from cognee.infrastructure.databases.graph import get_graph_engine
-from cognee.infrastructure.databases.graph.networkx.adapter import NetworkXAdapter
 from cognee.modules.retrieval.base_retriever import BaseRetriever
 from cognee.modules.retrieval.utils.completion import generate_completion
 from cognee.modules.retrieval.exceptions import SearchTypeNotSupported, CypherSearchError
@@ -31,8 +30,7 @@ class CypherSearchRetriever(BaseRetriever):
         """
         Retrieves relevant context using a cypher query.
 
-        If the graph engine is an instance of NetworkXAdapter, raises SearchTypeNotSupported. If
+        If any error occurs during execution, logs the error and raises CypherSearchError.
-        any error occurs during execution, logs the error and raises CypherSearchError.
 
         Parameters:
         -----------
@@ -46,12 +44,6 @@ class CypherSearchRetriever(BaseRetriever):
         """
         try:
             graph_engine = await get_graph_engine()
-
-            if isinstance(graph_engine, NetworkXAdapter):
-                raise SearchTypeNotSupported(
-                    "CYPHER search type not supported for NetworkXAdapter."
-                )
-
             result = await graph_engine.query(query)
         except Exception as e:
             logger.error("Failed to execture cypher search retrieval: %s", str(e))

cognee/modules/retrieval/exceptions/exceptions.py

@@ -1,8 +1,8 @@
 from fastapi import status
-from cognee.exceptions import CogneeApiError, CriticalError
+from cognee.exceptions import CogneeValidationError, CogneeSystemError
 
 
-class SearchTypeNotSupported(CogneeApiError):
+class SearchTypeNotSupported(CogneeValidationError):
     def __init__(
         self,
         message: str = "CYPHER search type not supported by the adapter.",
@@ -12,7 +12,7 @@ class SearchTypeNotSupported(CogneeApiError):
         super().__init__(message, name, status_code)
 
 
-class CypherSearchError(CogneeApiError):
+class CypherSearchError(CogneeSystemError):
     def __init__(
         self,
         message: str = "An error occurred during the execution of the Cypher query.",
@@ -22,11 +22,17 @@ class CypherSearchError(CogneeApiError):
         super().__init__(message, name, status_code)
 
 
-class NoDataError(CriticalError):
+class NoDataError(CogneeValidationError):
-    message: str = "No data found in the system, please add data first."
+    def __init__(
+        self,
+        message: str = "No data found in the system, please add data first.",
+        name: str = "NoDataError",
+        status_code: int = status.HTTP_404_NOT_FOUND,
+    ):
+        super().__init__(message, name, status_code)
 
 
-class CollectionDistancesNotFoundError(CogneeApiError):
+class CollectionDistancesNotFoundError(CogneeValidationError):
     def __init__(
         self,
         message: str = "No collection distances found for the given query.",

cognee/modules/retrieval/natural_language_retriever.py

@@ -1,7 +1,6 @@
 from typing import Any, Optional
 from cognee.shared.logging_utils import get_logger
 from cognee.infrastructure.databases.graph import get_graph_engine
-from cognee.infrastructure.databases.graph.networkx.adapter import NetworkXAdapter
 from cognee.infrastructure.llm.LLMGateway import LLMGateway
 from cognee.modules.retrieval.base_retriever import BaseRetriever
 from cognee.modules.retrieval.exceptions import SearchTypeNotSupported
@@ -123,9 +122,6 @@ class NaturalLanguageRetriever(BaseRetriever):
         """
         graph_engine = await get_graph_engine()
 
-        if isinstance(graph_engine, (NetworkXAdapter)):
-            raise SearchTypeNotSupported("Natural language search type not supported.")
-
         return await self._execute_cypher_query(query, graph_engine)
 
     async def get_completion(self, query: str, context: Optional[Any] = None) -> Any:

cognee/modules/search/exceptions/__init__.py

@@ -0,0 +1,7 @@
+"""
+Custom exceptions for the Cognee API.
+
+This module defines a set of exceptions for handling various data errors
+"""
+
+from .exceptions import UnsupportedSearchTypeError

cognee/modules/search/exceptions/exceptions.py

@@ -0,0 +1,15 @@
+from cognee.exceptions import (
+    CogneeValidationError,
+)
+from fastapi import status
+
+
+class UnsupportedSearchTypeError(CogneeValidationError):
+    def __init__(
+        self,
+        search_type: str,
+        name: str = "UnsupportedSearchTypeError",
+        status_code: int = status.HTTP_400_BAD_REQUEST,
+    ):
+        message = f"Unsupported search type: {search_type}"
+        super().__init__(message, name, status_code)

cognee/modules/search/methods/search.py

@@ -3,9 +3,8 @@ import json
 import asyncio
 from uuid import UUID
 from typing import Callable, List, Optional, Type, Union
-
+from cognee.modules.search.exceptions import UnsupportedSearchTypeError
 from cognee.context_global_variables import set_database_global_context_variables
-from cognee.exceptions import InvalidValueError
 from cognee.modules.retrieval.chunks_retriever import ChunksRetriever
 from cognee.modules.retrieval.insights_retriever import InsightsRetriever
 from cognee.modules.retrieval.summaries_retriever import SummariesRetriever
@@ -143,7 +142,7 @@ async def specific_search(
     search_task = search_tasks.get(query_type)
 
     if search_task is None:
-        raise InvalidValueError(message=f"Unsupported search type: {query_type}")
+        raise UnsupportedSearchTypeError(str(query_type))
 
     send_telemetry("cognee.search EXECUTION STARTED", user.id)
 

cognee/modules/settings/get_settings.py

@@ -88,8 +88,8 @@ def get_settings() -> SettingsDict:
                 "models": {
                     "openai": [
                         {
-                            "value": "gpt-4o-mini",
+                            "value": "gpt-5-mini",
-                            "label": "gpt-4o-mini",
+                            "label": "gpt-5-mini",
                         },
                         {
                             "value": "gpt-4o",

cognee/modules/users/exceptions/exceptions.py

@@ -1,8 +1,8 @@
-from cognee.exceptions import CogneeApiError
+from cognee.exceptions import CogneeValidationError
 from fastapi import status
 
 
-class RoleNotFoundError(CogneeApiError):
+class RoleNotFoundError(CogneeValidationError):
     """User group not found"""
 
     def __init__(
@@ -14,7 +14,7 @@ class RoleNotFoundError(CogneeApiError):
         super().__init__(message, name, status_code)
 
 
-class TenantNotFoundError(CogneeApiError):
+class TenantNotFoundError(CogneeValidationError):
     """User group not found"""
 
     def __init__(
@@ -26,7 +26,7 @@ class TenantNotFoundError(CogneeApiError):
         super().__init__(message, name, status_code)
 
 
-class UserNotFoundError(CogneeApiError):
+class UserNotFoundError(CogneeValidationError):
     """User not found"""
 
     def __init__(
@@ -38,7 +38,7 @@ class UserNotFoundError(CogneeApiError):
         super().__init__(message, name, status_code)
 
 
-class PermissionDeniedError(CogneeApiError):
+class PermissionDeniedError(CogneeValidationError):
     def __init__(
         self,
         message: str = "User does not have permission on documents.",
@@ -48,7 +48,7 @@ class PermissionDeniedError(CogneeApiError):
         super().__init__(message, name, status_code)
 
 
-class PermissionNotFoundError(CogneeApiError):
+class PermissionNotFoundError(CogneeValidationError):
     def __init__(
         self,
         message: str = "Permission type does not exist.",

cognee/shared/exceptions/exceptions.py

@@ -1,8 +1,8 @@
-from cognee.exceptions import CogneeApiError
+from cognee.exceptions import CogneeValidationError
 from fastapi import status
 
 
-class IngestionError(CogneeApiError):
+class IngestionError(CogneeValidationError):
     def __init__(
         self,
         message: str = "Failed to load data.",

cognee/shared/utils.py

@@ -3,7 +3,6 @@
 import os
 import requests
 from datetime import datetime, timezone
-import networkx as nx
 import matplotlib.pyplot as plt
 import http.server
 import socketserver

cognee/tasks/completion/exceptions/exceptions.py

@@ -1,11 +1,11 @@
-from cognee.exceptions import CogneeApiError
+from cognee.exceptions import CogneeValidationError
 from fastapi import status
 
 
-class NoRelevantDataError(CogneeApiError):
+class NoRelevantDataError(CogneeValidationError):
     """
     Represents an error when no relevant data is found during a search. This class is a
-    subclass of CogneeApiError.
+    subclass of CogneeValidationError.
 
     Public methods:
 

cognee/tasks/documents/classify_documents.py

@@ -10,6 +10,7 @@ from cognee.modules.data.processing.document_types import (
 )
 from cognee.modules.engine.models.node_set import NodeSet
 from cognee.modules.engine.utils.generate_node_id import generate_node_id
+from cognee.tasks.documents.exceptions import WrongDataDocumentInputError
 
 EXTENSION_TO_DOCUMENT_CLASS = {
     "pdf": PdfDocument,  # Text documents
@@ -111,6 +112,9 @@ async def classify_documents(data_documents: list[Data]) -> list[Document]:
         - list[Document]: A list of Document objects created based on the classified data
           documents.
     """
+    if not isinstance(data_documents, list):
+        raise WrongDataDocumentInputError("data_documents")
+
     documents = []
     for data_item in data_documents:
         document = EXTENSION_TO_DOCUMENT_CLASS[data_item.extension](

cognee/tasks/documents/exceptions/__init__.py

@@ -0,0 +1,11 @@
+"""
+Custom exceptions for the Cognee API.
+
+This module defines a set of exceptions for handling various data errors
+"""
+
+from .exceptions import (
+    WrongDataDocumentInputError,
+    InvalidChunkSizeError,
+    InvalidChunkerError,
+)

cognee/tasks/documents/exceptions/exceptions.py

@@ -0,0 +1,36 @@
+from cognee.exceptions import (
+    CogneeValidationError,
+    CogneeConfigurationError,
+)
+from fastapi import status
+
+
+class WrongDataDocumentInputError(CogneeValidationError):
+    """Raised when a wrong data document is provided."""
+
+    def __init__(
+        self,
+        field: str,
+        name: str = "WrongDataDocumentInputError",
+        status_code: int = status.HTTP_422_UNPROCESSABLE_ENTITY,
+    ):
+        message = f"Missing of invalid parameter: '{field}'."
+        super().__init__(message, name, status_code)
+
+
+class InvalidChunkSizeError(CogneeValidationError):
+    def __init__(self, value):
+        super().__init__(
+            message=f"max_chunk_size must be a positive integer (got {value}).",
+            name="InvalidChunkSizeError",
+            status_code=status.HTTP_400_BAD_REQUEST,
+        )
+
+
+class InvalidChunkerError(CogneeValidationError):
+    def __init__(self):
+        super().__init__(
+            message="chunker must be a valid Chunker class.",
+            name="InvalidChunkerError",
+            status_code=status.HTTP_400_BAD_REQUEST,
+        )

cognee/tasks/documents/extract_chunks_from_documents.py

@@ -8,6 +8,7 @@ from cognee.modules.data.models import Data
 from cognee.infrastructure.databases.relational import get_relational_engine
 from cognee.modules.chunking.TextChunker import TextChunker
 from cognee.modules.chunking.Chunker import Chunker
+from cognee.tasks.documents.exceptions import InvalidChunkSizeError, InvalidChunkerError
 
 
 async def update_document_token_count(document_id: UUID, token_count: int) -> None:
@@ -37,6 +38,13 @@ async def extract_chunks_from_documents(
         - The `read` method of the `Document` class must be implemented to support the chunking operation.
         - The `chunker` parameter determines the chunking logic and should align with the document type.
     """
+    if not isinstance(max_chunk_size, int) or max_chunk_size <= 0:
+        raise InvalidChunkSizeError(max_chunk_size)
+    if not isinstance(chunker, type):
+        raise InvalidChunkerError()
+    if not hasattr(chunker, "read"):
+        raise InvalidChunkerError()
+
     for document in documents:
         document_token_count = 0
 
@@ -48,5 +56,3 @@ async def extract_chunks_from_documents(
             yield document_chunk
 
         await update_document_token_count(document.id, document_token_count)
-
-        # todo rita

cognee/tasks/graph/exceptions/__init__.py

@@ -0,0 +1,12 @@
+"""
+Custom exceptions for the Cognee API.
+
+This module defines a set of exceptions for handling various data errors
+"""
+
+from .exceptions import (
+    InvalidDataChunksError,
+    InvalidGraphModelError,
+    InvalidOntologyAdapterError,
+    InvalidChunkGraphInputError,
+)

cognee/tasks/graph/exceptions/exceptions.py

@@ -0,0 +1,41 @@
+from cognee.exceptions import (
+    CogneeValidationError,
+    CogneeConfigurationError,
+)
+from fastapi import status
+
+
+class InvalidDataChunksError(CogneeValidationError):
+    def __init__(self, detail: str):
+        super().__init__(
+            message=f"Invalid data_chunks: {detail}",
+            name="InvalidDataChunksError",
+            status_code=status.HTTP_400_BAD_REQUEST,
+        )
+
+
+class InvalidGraphModelError(CogneeValidationError):
+    def __init__(self, got):
+        super().__init__(
+            message=f"graph_model must be a subclass of BaseModel (got {got}).",
+            name="InvalidGraphModelError",
+            status_code=status.HTTP_400_BAD_REQUEST,
+        )
+
+
+class InvalidOntologyAdapterError(CogneeConfigurationError):
+    def __init__(self, got):
+        super().__init__(
+            message=f"ontology_adapter lacks required interface (got {got}).",
+            name="InvalidOntologyAdapterError",
+            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
+        )
+
+
+class InvalidChunkGraphInputError(CogneeValidationError):
+    def __init__(self, detail: str):
+        super().__init__(
+            message=f"Invalid chunk inputs or LLM Chunkgraphs: {detail}",
+            name="InvalidChunkGraphInputError",
+            status_code=status.HTTP_400_BAD_REQUEST,
+        )

cognee/tasks/graph/extract_graph_from_data.py

@@ -12,6 +12,12 @@ from cognee.modules.graph.utils import (
 )
 from cognee.shared.data_models import KnowledgeGraph
 from cognee.infrastructure.llm.LLMGateway import LLMGateway
+from cognee.tasks.graph.exceptions import (
+    InvalidGraphModelError,
+    InvalidDataChunksError,
+    InvalidChunkGraphInputError,
+    InvalidOntologyAdapterError,
+)
 
 
 async def integrate_chunk_graphs(
@@ -21,6 +27,20 @@ async def integrate_chunk_graphs(
     ontology_adapter: OntologyResolver,
 ) -> List[DocumentChunk]:
     """Updates DocumentChunk objects, integrates data points and edges into databases."""
+
+    if not isinstance(data_chunks, list) or not isinstance(chunk_graphs, list):
+        raise InvalidChunkGraphInputError("data_chunks and chunk_graphs must be lists.")
+    if len(data_chunks) != len(chunk_graphs):
+        raise InvalidChunkGraphInputError(
+            f"length mismatch: {len(data_chunks)} chunks vs {len(chunk_graphs)} graphs."
+        )
+    if not isinstance(graph_model, type) or not issubclass(graph_model, BaseModel):
+        raise InvalidGraphModelError(graph_model)
+    if ontology_adapter is None or not hasattr(ontology_adapter, "get_subgraph"):
+        raise InvalidOntologyAdapterError(
+            type(ontology_adapter).__name__ if ontology_adapter else "None"
+        )
+
     graph_engine = await get_graph_engine()
 
     if graph_model is not KnowledgeGraph:
@@ -55,6 +75,14 @@ async def extract_graph_from_data(
     """
     Extracts and integrates a knowledge graph from the text content of document chunks using a specified graph model.
     """
+
+    if not isinstance(data_chunks, list) or not data_chunks:
+        raise InvalidDataChunksError("must be a non-empty list of DocumentChunk.")
+    if not all(hasattr(c, "text") for c in data_chunks):
+        raise InvalidDataChunksError("each chunk must have a 'text' attribute")
+    if not isinstance(graph_model, type) or not issubclass(graph_model, BaseModel):
+        raise InvalidGraphModelError(graph_model)
+
     chunk_graphs = await asyncio.gather(
         *[LLMGateway.extract_content_graph(chunk.text, graph_model) for chunk in data_chunks]
     )

cognee/tasks/ingestion/exceptions/__init__.py

@@ -0,0 +1,8 @@
+"""
+Custom exceptions for the Cognee API.
+
+This module defines a set of exceptions for handling various application errors,
+such as System, Validation, Configuration or TransientErrors
+"""
+
+from .exceptions import S3FileSystemNotFoundError

cognee/tasks/ingestion/exceptions/exceptions.py

@@ -0,0 +1,12 @@
+from cognee.exceptions import CogneeSystemError
+from fastapi import status
+
+
+class S3FileSystemNotFoundError(CogneeSystemError):
+    def __init__(
+        self,
+        name: str = "S3FileSystemNotFoundError",
+        status_code: int = status.HTTP_500_INTERNAL_SERVER_ERROR,
+    ):
+        message = "Could not find S3FileSystem."
+        super().__init__(message, name, status_code)